Podcasts about u s

In episode 2020, Jack and Miles are joined by comedian and host of Go Fact Yourself, J. Keith van Straaten, to discuss… Pete Hegseth’s Pentagon Is Spending Billions On War... And Millions On Lobster, Trump Playing 4D Checkers Baby With Save Act, Old Habits…Iran/Midterms, Is Hollywood’s UFO Trend A Government PSYOP? And more! Is the Iran war really costing the US $2bn per day? Pentagon Should Focus on Defense Priorities, not Lavish Dinners, After Historic $93.4B “Use-It-or-Lose-It” September Pete Hegseth Blew Billions on Fruit Basket Stands, Chairs, and Crab Hollywood Is Suddenly Taking UFOs Seriously, With Rival “Disclosure” Projects in the Works (Exclusive) ‘A lot of stories but very few facts’: sceptics push back on buzzy UFO documentary MAGA Congresswoman Claims UFOs Might Be ‘Interdimensional Beings’ UFOs, Aliens & Steven Spielberg's 20-Year Obsession Close Encounters: Cultural Impact Claim: NASA tried to stop Spielberg's 'Close Encounters of the Third Kind' CIA Influence on 'The Day the Earth Stood Still' The Day the Earth Stood Still: Rejected by the US Air Force, but aided by the CIA? LISTEN: blackbird by Victoria CanalSee omnystudio.com/listener for privacy information.

In this episode of Gangland Wire, I sit down with retired FBI agent Geoff Kelly, a specialist in art theft investigations who inherited one of the most notorious unsolved cases in American history—the 1990 robbery at the Isabella Stewart Gardner Museum. He recently wrote a book about this theft titled 13 Perfect Fugitives: The True Story of Mob, Murder, and the World’s Largest Art Heist. Kelly's law enforcement career began as a New York City transit police officer before transitioning to the FBI. Like many agents, he initially sought violent crime work. Instead, he was assigned to economic crimes before eventually transferring to a violent crime squad. It was there that he encountered the Gardner case—a cold case largely untouched by senior agents at the time. The robbery itself remains extraordinary: two men posing as police officers gained entry to the museum and stole 13 works of art, including masterpieces by Rembrandt. More than three decades later, none of the works have been recovered. Inside the Gardner Heist Geoff explains how art theft is often misunderstood. Popular culture portrays refined, sophisticated criminals orchestrating elaborate capers. The reality, he says, is usually more opportunistic and frequently violent. Art theft often intersects with organized crime, drug trafficking, and even homicide. Massachusetts has a documented history of art-related crimes, and several individuals connected to the Gardner investigation met violent ends. The criminal underworld surrounding stolen art is less about wealthy collectors hiding paintings in private vaults and more about leverage—using artwork as collateral in criminal negotiations. The FBI's Art Crime Evolution Following the 2003 looting of Iraq's National Museum during the Baghdad invasion, the FBI formalized its Art Crime Team. Kelly discusses how intelligence gathering, informants, and international cooperation became central tools in recovering stolen artifacts. He emphasizes that solving art crimes often depends less on forensic breakthroughs and more on human intelligence. Informants remain essential, especially in cases where organized crime overlaps with high-value theft. Kelly also discusses his upcoming book, 13 Perfect Fugitives, which explores the intersections of mobsters, murder, and the illicit art market. Organized Crime and the Reality of Stolen Art Drawing on my own experience working organized crime in Kansas City, I found clear parallels between traditional mob rackets and art theft networks. The same structures—intimidation, secrecy, and violence—apply. Once a painting disappears into criminal circulation, it becomes a liability as much as an asset. Kelly challenges the myth that thieves profit easily from masterpieces. High-profile works are difficult to sell. The black-market art world is volatile and dangerous. In many cases, the artwork becomes bargaining collateral rather than a cash windfall. A Case Still Waiting for Closure More than 30 years later, the Gardner Museum still displays empty frames where the paintings once hung. Kelly remains committed to the idea that public awareness may eventually generate new leads.  The Gardner heist stands as both a cultural tragedy and a criminal mystery—one that continues to intersect with organized crime, violence, and international intrigue. Hit me up on Venmo for a cup of coffee or a shot and a beer @ganglandwire Click here to “buy me a cup of coffee” Subscribe to the website for weekly notifications about updates and other Mob information. To go to the store or make a donation or rent Ballot Theft: Burglary, Murder, Coverup, click here To rent ‘Brothers against Brothers’ or ‘Gangland Wire,’ the documentaries click here.  To purchase one of my books, click here. Transcript [0:00] Hey, you guys, Gary Jenkins back here in studio Gangland Wire. Y’all know me. I’m a retired Kansas City Police Intelligence Unit detective and now podcaster and documentary filmmaker. I have in the studio today… Jeff Kelly, he’s a now-retired FBI agent. He was an expert in recovering stolen artifacts and art pieces. He was involved. He wasn’t involved in the original theft of the Boston art theft, the Isabella Stewart Gardner Museum, but he ended up inheriting that case. So welcome, Jeff. Hi. Thanks, Gary. Nice to be here. And guys, I need to mention this right off the bat. Jeff has a book, 13 Perfect Fugitives, The True Story of the Mob, Murder, and the World’s Largest Art Heist. Be out on Amazon. I’ll have links down below in the show notes if you want to get that book. I think it would be pretty interesting. I was telling Jeff, I just interviewed Joe Ford, the million-dollar detective, the guy that goes after classic cars, and I read that book. I love these kind of caper kind of books and caper crimes. Those are the ones I like the best is the caper crimes. And Jeff is an expert at working caper crimes. And that’s what these are, capers. So Jeff, how did you get into this? Now you came on the FBI. You were a policeman before, I believe. So tell the guys a little bit about yourself and your FBI career. Yeah, I started out with the Metropolitan Transportation Authority Police in New York City. It was a transit cop. I did that for three years. And then I got into the FBI in October of 95. [1:30] And my goal was always, I wanted to work violent crime. That’s what drew me to law enforcement in the first place, working bank robberies and kidnappings and fugitives. I had to do my five years on working economic crime, telemarketing fraud. It was interesting, but not all that exciting. And finally in 2000, I got my transfer to the violent crime squad. And I loved working it. And I did it for my entire career from then on, right up until my retirement in 2024. But back then, art theft was considered a major theft violation, [2:01] and it was worked by the Violent Crime Squad. And so in 2002… My supervisor dumped this old moribund cold case in my lap. It was the Isabella Stewart Gardner Museum heist. [2:15] Nobody wanted it on the squad, so they figured, let’s give it to the new guy. I was ecstatic to get it because I’d heard about it. I went to school in Boston. I went to Boston University and graduated the year before it happened, but I knew about it. [2:28] That’s how I started working this case, this particular case, and then the following year during the U.S., there was a, the U.S. And coalition forces invaded Baghdad in Iraq. And during a 36-hour period, more than 15,000 objects of very, very important cultural history were looted from the National Museum of Iraq. And it’s really one of the most important museums in the world in terms of our shared history. Kind of the cradle of civilization over there in the Tigers and Euphrates River. Yeah, and that was the time when the FBI kind of belatedly realized that there was no art crime team to investigate this. And of course, FBI agents have been working art theft like any other property crime since the beginning of the FBI’s existence, but there was no codified team. So they did a canvas for the team in 2004 and I applied for it because at this point I’d been working the Gardner case for a couple of years and really was fascinated by it and made the team. And so then over the next 20 years, we continued to expand the team both in size and in scope and in our intelligence base and knowledge base. And when I left the Bureau in 2024, it was and still is a tremendous team with a lot of very dedicated and professional agents and professional support. [3:51] Now, guys, if you don’t know about the Isabella Stewart Gardner case, there was a Netflix documentary on it a few years ago. It was an art museum in Boston. [4:01] Two guys showed up. They had Boston police uniforms on, and they got in. They basically, it was an armed robbery, and they took control of the museum. The guards were in there late at night and took these really valuable paintings out. I believe you told me earlier they were Remington paintings. We’ll get into that. And it was a violent crime. It was an armed robbery of paintings, and you told me about other armed robberies of paintings. I think you got into some other armed robberies of paintings. You always think of, as you mentioned before, the Thomas Crown Affair character that goes out and does these sophisticated art thefts. That’s not always true, is it? It’s never that way, but it doesn’t matter. Don’t let the facts get in the way of a good story. Everybody wants to believe that art thefts are pulled off by the Thomas Crown Affairs and these gentlemen thieves repel in through skylights and do all that fancy stuff, put it in their underground lair. That’s just not the way it works. But if you look to art theft. [4:55] Massachusetts really is a cradle of art theft in this country, and it’s very unique. The first armed robbery of a museum occurred in Boston in 1972. It was committed by a guy named Al Monday, who was a prolific art thief. And they stole four pieces from the Worcester Art Museum in central Massachusetts with a gun. They ended up shooting the guard. And one of the pieces that they stole was a Rembrandt called St. Bartholomew. [5:26] And in keeping with the milieu of true art thieves, the paintings were stored on a pig farm just over the state line in Rhode Island. And when this Connecticut safecracker by the name of Chucky Carlo, who was looking at some serious time in prison for some of the crimes that he committed, when he found out that Al Monday had these paintings, he just simply kidnapped Al Monday and stuck a gun in his ribs and said he would kill him if he didn’t give him the paintings. which is no honor among thieves. And Al turned over the paintings, Chucky returned them, and he got a very significant break on his pending jail sentence. Right here in 1972, Boston thieves see Rembrandt as a valuable get-out-of-jail-free card. [6:09] And then if we jump forward three years to 1975, there was a very skilled art thief, really a master thief by the name of Miles Conner. I interviewed Miles for my book. It was very gracious of him to sit down with me for it. And he had robbed or committed a burglary of the Woolworth estate up in Maine, the family, the five and dime family magnets. And he got caught for it because he tried to sell those paintings to an undercover FBI agent. And so he was looking at 12 years in prison for it. And he was out on bail. And he reached out to a family friend who was a state trooper. And he asked him, how can I get away with this one? How can I get out of this? Because he was in serious trouble. The trooper’s response was meant to be hyperbolic. The trooper said, Miles, it’s going to take you a Rembrandt to get out of this one. [6:57] And so Miles said, okay, I’ll go get a Rembrandt. And he got a crew together and they did a daylight smash and grab at the Museum of Fine Arts in Boston, just across the street from the Gardner. And they stole Rembrandt, the girl in a gold-trimmed cloak. [7:12] And he was able to return that painting. Instead of doing 12 years, he did 28 months. And he even managed to, he told me he even managed to get the $10,000 reward in the process. So you have this atmosphere in Massachusetts that Rembrandts are a valuable commodity, right? They can help you out in a jam. And so I think it’s no coincidence that in 1990, when the Gardner Museum heist came down, the Gardner Museum had this array of motion sensors all throughout the museum. It would alert to wherever you went, every gallery, hallway, whatever. [7:49] And we know from these motion sensors that after, as you said, the two guys went in disguised as cops and bluffed their way into the museum, they made a beeline for the Dutch room, which is the room of all things Rembrandt. They stole three Rembrandts. They stole a fourth piece called Landscape with an Obelisk, which was actually by Govard Flink, but it had been misattributed to Rembrandt until the mid 80s. And then they took a large Rembrandt oil-on-panel off the wall and it was recovered the next morning leaning against a piece of furniture. We believe they just overlooked it in the dark. So out of the 13 pieces taken, three were Rembrandt, a fourth was misattributed to Rembrandt, and there was going to be a 14th piece taken, which was also Rembrandt. It definitely falls into that theory that this was going to be a hold-on to these pieces for a while and see if you can use them for a break. [8:48] Interesting. Now, back in the 70s, for example, when somebody would work in an art robbery like that or an art theft, you got your tried and true ways of working a crime. You got to have sources, you got to have witnesses, and hopefully you can get a crime like this. You can get a source that says, hey, this guy, we had a guy in Kansas City that he was a fence for these kinds of guys. He had an antique auction and he took all this stuff and got it somewhere else. So at the time, just use your regular police methods. And what changed over the years as you’ve done this? Yeah, certainly we’ve become much more sophisticated with the techniques that we use. But at the end of the day, it’s always still going to be intelligence. But I found from working my entire career in violent crime, virtually my whole career in violent crime, the sources are crucial. Having a good informant can make and break a case. And working art theft investigations, you’re certainly going to have the same types of fences of informants, fences for stolen property and what they’re hearing about what organized crime guys are doing and what drug guys are doing. But it also opened up a whole new avenue of sources for me as working in art investigations, because now you’ve got pawn shops and gallery owners and auction houses, and they’re in a position to know when not only when stolen artwork is coming in, but also fakes and forgeries. We spoke about this, that. [10:16] Somebody comes in with one valuable piece that would be very difficult for somebody in his or her position to come across one piece like this, let alone a dozen of them. That really points to probably a fake. And so that’s really the key to solving these things is just having a good intelligence base who’s going to let us know about when something comes up that’s either stolen or it’s been forged. [10:43] Brings up a question. In my mind, did you ever work a gallery owner or a gallery [10:48] that then would filter in, knowingly filter in some fakes every once in a while? They couldn’t do it 100% of the time, but you could certainly make some extra money by filtering fakes out of it because many people would get it and they’d never know. Nobody would ever know. Listen, it is a really difficult thing when you’re working these types of crimes because unlike bank robber, you go into a bank and you stick them up with a gun and take them on. It’s not up to the government to be able to prove at trial that you knew that the bank was insured by the FDIC. You went in and you robbed it, you committed the offense. When you’re talking about interstate transportation of stolen property or possession of stolen property, there are what’s called specific intent crimes, meaning you have to prove the element of knowledge. You have to be able to prove that the person knew that that item was stolen. Not that it said it was stolen. and you had to show that they knew it. And that’s a really high hurdle to overcome. And typically what we do to try and prove that specific intent is we’re going to go through. [11:53] Recorded statements made to a source or to an undercover or emails or texts or something that we can show that this person knew that item was stolen. And so we would see that a lot in auction houses and galleries. There’s a lot of willful blindness where a lot of gallery owners and auction houses, they’re going to look the other way because it’s too lucrative to pass up. And in fact, in 2015, the art crime team, once we received information that ISIL or ISIS was using looted cultural property from Syria and Iraq as a form, a viable form of terrorism financing. And we put auction houses and gallery owners on notice in 2015, and we basically told them that if you’re selling objects of cultural patrimony or cultural heritage with a dubious provenance, like a wink and a nod, you may be unwittingly or wittingly funding terrorism. While we never charged anybody with it, hopefully it was an eye-opener that when you’re getting into this world, it’s not a victimless crime. There are very real victims involved. [13:07] And that’s one of the things that really is interesting about working our crime investigations. And I used to get ribbed by my friends who were not on the art crime team about [13:18] where like the wine and cheese squad were raised and everything. But our subjects are far from it. We’re dealing with organized crime, gangs, terrorists. This is no joke. These are serious individuals and the stakes are high. And in the Gardner case, three or four people that we believe were involved in the heist were murdered a year after the Gardner case crime occurred. Yeah, I was just going to go back to that a little bit, as we said before, a little bit like the Lufthansa case. All of a sudden, everybody that was involved in the theft. Started dropping like flies. So tell the guys about that. That is really interesting. [14:00] Yeah. So the two individuals that we believe went into the museum dressed as cops, just a week shy of the one-year anniversary, one of the guys was found dead in his apartment of an acute overdose of cocaine, intravenous. And his family admitted that he used Coke, but they said he was terrified of needles. He was scared of needles. So it really looked to be like a hotshot, an intentional overdose of cocaine. Two weeks later, the other guy who we believe went into the museum with him, his wife reported him missing. And a couple of weeks later, his bullet riddled body was recovered in the trunk of his car out by Logan Airport in East Boston. There was another member of that crew. These were all part of the same crew. This Carmelo Merlino, who was a Boston mobster, had an auto shop down in the Dorchester section of Boston. Another member of his crew, a guy named Bobby, six weeks after the heist, he brought in, he visited a jeweler in the downtown crossing jewelry district in Boston. He came in with this object and he unwrapped it. It was an eagle. [15:03] It was the finial from the Napoleonic flag that was stolen in the Gardner heist. And he asked the jeweler, how much is this thing worth? And the jeweler looked at it and he said, it’s worth nothing. Because he immediately recognized it as one of the people that had been stolen six weeks earlier from the Gardner heist. And then a few months later, Bobby was stabbed to death and nearly decapitated on the front porch of his house. And the responding police saw that his house had been broken into and ransacked like his killers had been looking for something. There was a fourth guy, Jimmy, who bragged to his girlfriend a few months after the heist that he had a couple of pieces from the Gardner Museum hidden in his attic. [15:47] And in February of 1990, 11 months after the heist, he was executed on his front porch in what the local police called a mob hit. So, yeah, these are the types of crimes that have a tendency to have a chilling effect on anybody who harbors any aspirations to come forward with information. Yeah, and we talked earlier a little bit about, like, the crime itself, and the statute of limitations is up on that, what you said, and the crime itself, but how we talked a little bit and explained to them about how this could be part of a RICO case. And you’ve got the murders and you’ve got the actual theft and whatever they did with the paintings, then maybe you could get over after a Bob boss as a Rico case. Tell the guys a little bit about doing that. Yeah. [16:32] I’ve heard it so many times in more than two decades that I worked the case and people would say, geez, why don’t people come forward? They’re just paintings. There are so many times they’re just paintings. They’re like, yeah, they are, but there’s two things about that. Number one, there’s some dead bodies on these paintings, three or four, and that there’s no statute of limitations for murder. And so if you implicate yourself in the theft or you implicate yourself in possessing or transporting these paintings at any time, the fear is that you’re then implicating yourself in a homicide. And the other aspect of this, which I think has a chilling effect, is the fact that transportation of stolen property is one of the predicate acts for RICO, racketeering influence corrupt organization case. And RICO is basically, Gary, is basically an entire organization is corrupt. Yeah. There’s no legitimate purpose. It’s what we think about the mob and the [17:27] FBI has taken down the mob in the past. So if you implicate yourself in stolen property and you’re part of organized crime, that’s one of the predicate acts for a RICO. And that’s basically life sentences. And so one of my goals in the years and years that I worked in this case was to try and convince people that you could come forward with information and the U S attorney’s offices, whether it’s up in Boston or new Haven or Philadelphia. [17:58] Would be willing to figure out a way to get the paintings back with immunity from prosecution for a RICO case. Look, that’s a high hurdle. That’s a high hurdle to convince somebody that if you come forward, you’re not going to get charged and you’re eligible for millions of dollars in reward. That’s a tough bill to swallow, but it’s the truth. I’m retired from the FBI now. I can tell you that it was, it’s a, it was, and still is a bona fide offer. And that’s one of the goals that I’ve always tried to impress on anyone is the opportunity to become a millionaire without going to jail. There you go, Jeff. Can you, now you’re not with the Bureau anymore. Can you go out, if you could go out and find them and bring them in, could you collect that reward? I would certainly hope so. [18:48] I can’t tell you how many of my friends thought that I had some of these paintings stashed in my basement. Waiting for retirement to go turn them in the next day. I think half the guys I worked with were expecting to see me pull into the parking lot of the FBI. [19:01] Big package, but no. But yeah, I suppose I could. By this point, I can tell you the amount of my very being that I put into this case over two days. Yeah. I just would love to see these paintings go back just because they need to be back at the museum. That’s where they belong. Now, these crimes, they seem, You said there’s a lot of murders attached to this. They seem a little boring. Did you have any exciting moments trying to pop anybody or do any surveillances? I know we did a big surveillance of a bunch of junkies that were going around stealing from small museums around the Midwest. And we follow them here in Kansas City. And they would have been pretty exciting had we had a confrontation with them. Did you have any exciting moments in this? It actually was a fascinating case. And for the first, there’s the really boring aspects of this job and tedious aspects. And I would say that in my, two decades working this case, I probably did, I don’t know, 50, 60, 70 consent searches, searching in attics and basements and crawling through crawl spaces and just getting sweaty and covered in cobwebs. But the break in the case for me came in 2009 when one of the guys who was part of Merlino’s crew who was deceased, his niece came forward to me and told me that the paintings. Some of them had been hidden up in this guy’s hide at his house up in Maine. I went up to Maine with Anthony Amore, who’s the director of security for the Gardner Museum. We worked on this case together for years. [20:29] And then we found that hide. And then we interviewed, right from there, we went and interviewed Guarenti. That’s the guy, Bobby Guarenti. We interviewed his widow and she broke down and admitted that he once showed her the paintings and she gave them to a guy down in Connecticut. And we identified that guy and we interviewed him. My name is Bobby Gentile. He’s a made member of the Philly Mob. He got straightened out with his crew back in the late 90s. [20:54] And he refused to cooperate. And then that’s where we really just started getting, using a lot of ingenuity to try and break it. And an agent down in the New Haven office, a guy by the name of Jamie Lawton, he joined our team and we started working this case. And he had a source who knew Gentile, Bobby Gentile, and the source started buying drugs from Gentile. Ah, there we go. We ended up arresting Gentile and we did a search warrant at his house. And it was crazy. Like we recovered, I want to say seven handguns, loaded handguns lying all over the place. He had a pump action shotgun hanging by the front door. He had high explosives. We had to evacuate the house and call him the bomb squad. But the interesting thing was he had the March 19th, 1990 edition of the Boston Herald with headlines about the Gardner heist and tucked inside that newspaper was a handwritten list of all the stolen items. With what looked like their black market values. This is in the house of a guy who swore up and down that he’d never heard of the Gardner Museum. And we were able to figure out who wrote the list. It was written by none other than Al Monday, who’s the guy that did the first armed robbery of a museum, of a Rembrandt. And we interviewed him and he told us that he wrote that list for Bobby Gentile and his buddy up in Maine, Bobby Garanti, because they had a buyer for the paintings and they wanted to know what they were worth. [22:24] So yeah, and then Gentile took 30 months. [22:28] He wouldn’t cooperate. And while he was incarcerated, we turned two of his closest friends to becoming sources. And so when he got out of prison in February or April of 2014, they started talking to him and talked about the gardener and they said they might know somebody who’d want to buy him. That’s how we then introduced an undercover agent. Gentile was introduced to Tony, this undercover FBI agent. Over six months, they had long talks about selling the paintings. Unfortunately, before Gentile would sell the paintings, he wanted to do a drug deal first, which we couldn’t allow to happen. We can’t let drugs walk on the street. So we had to take it down. And although we’d seized all these guns from Gentile back in 2012, he told the sources the FBI didn’t get all of his guns. Because of that disturbing comment, one of the sources asked Gentile if he could buy a gun for him. And Gentile sold him a loaded 38. So we arrested him again. And he still refused to cooperate. I don’t respect what he did for a living or a lot of the things that he did, but you do have to respect his adherence to his values. However, misguided they may have been, he took the code of omerta, the code of silence to heart, and he took it to his grave. He died, I think, in 2021 after going to prison a second time. [23:50] While we never got any paintings back, it was a tremendous ride, and I’m confident they will come back. It’s just going to be a question of when. Yeah, that kind of brings up the question that you hear people speculate. Did you ever run across this? Is there actually any rich old guys or an Arab sheik or somebody that buys stuff like this and then really keeps it and never shows it to anybody? Does that unicorn really exist? everybody wants that to be true i know virtually it’s not yeah there’s there’s never been a case of some wealthy what we call the doctor no theory some some reclusive billionaire with his underground lair filled with all the illicit stolen treasures of the world yeah that’s it’s never happened yeah i guess you never say never but but no look the majority statistically about three-quarters of everyone that collects art in this country does it for, and I assume it’s probably worldwide, does it for the investment potential. There’s a lot of money to be made in collecting art. It rarely, if ever, drops in value. So that’s why people collect art. If there’s somebody who has a particular piece that they want so badly that they’re going to commission its theft, it’s more the stuff of Hollywood. It could happen, but we’ve never seen that happen yet. Interesting. [25:14] We did have one case here where we had a medical doctor and he had it on the wall of his house. And it was, I believe it was a Western artist named Remington that these junkies stole out of Omaha. But it was such a minor piece that he could show it to anybody and they wouldn’t. They would say, oh, that’s cool. You got a Remington. [25:30] There’s plenty of those around. And he could afford a real deal Remington anyhow. So it wasn’t that big a deal. And that’s really what it comes down to is that art, high-end art does get stolen. It gets stolen quite often. The art market is about $60 billion, and the FBI, we estimated about $6 to $8 billion of that is illicit, whether it’s theft or fakes and forgeries. It’s a tremendous market, but it’s mostly second and third tier items. [26:02] Really valuable, well-known pieces. They do get stolen, but that’s the easy part. The easy part is stealing it. The hard part is monetizing it. That’s why you very rarely see recidivism among art thieves, high-end art thieves, because you do it once, and now you’re stuck with the thing. It’s easier to steal something else. You got to go out and boost fur coats and stuff to make a living. Exactly. Do a jewelry store robbery down there and make a living. And that’s exactly the point. That’s why you’re seeing a sea change in terms of art thefts, museum thefts. The Louvre was a great example of that. Dresden green vault robbery where 100 million euros in gems were stolen back in 2019 yeah. [26:45] Gems and jewelry, it can be broken down. It’s going to greatly diminish their value, but you can recut a gem. You can melt down the setting. You can monetize it for a greatly diminished value, but at least you can monetize it. You can’t cut up a Rembrandt into smaller pieces. [27:02] It’s only valuable as a whole complete piece. Yeah. I’m just thinking about that. We got a couple of guys, Jerry Scalise and Art Rachel in Chicago, flew to London, robbed a really valuable piece, the Lady Churchill’s diamond or something, I don’t remember, but really valuable piece and mailed it to somebody on their way to the airport and then got caught when they got back to Chicago and brought back to London and did 14 years in England and they never gave up that piece and nobody could, it never appeared anywhere, but it was just cut up and they didn’t make hardly any money off of it. Yeah. Look, there’s a, there’s much more profitable ways to. Yeah. To make an illicit living than stealing high-end artwork, but it does still get stolen. And that’s one of the cruel ironies when you’re talking about art theft is if somebody has a $20,000 piece of jewelry or a very expensive watch, they’re most likely going to lock it up in a safe in their bedroom or something. But you have a $10 million piece of artwork, you probably got it on the mantle. You’ve got it over the fireplace or in the front foyer of your house and probably doesn’t have a passive alarm system protecting it or security screws to keep it from being taken off the wall because people want to show it off. Yeah. It’s way too enticing. [28:24] Really? So, yes, you need to keep the word out there and keep this in people’s minds. And I’m sure the museum tries to do this in some ways in order, hopefully, that maybe somebody will say, oh. Yeah. [28:38] I think I saw that somewhere in this news program or on this podcast. [28:42] I’ll put some pictures on the podcast when I end up editing this. No, please do, Kerry. And that’s the thing. That’s the basis for the title of my book is it really is a fugitive investigation. And that’s how I work this case is fugitives and perfect fugitives because they’re not like their human counterparts. They’re not going to get tripped up on the silly things that we need to do as human beings, getting a driver’s license or whatnot. Yeah. [29:09] And so that’s how I worked the case. The FBI was really, I was always impressed with the FBI’s support that they gave me on this investigation. We did billboard campaigns and social media and a lot of things to get these images out there to the public, hoping it might resonate with somebody. And that’s really my goal for this book. I felt it should be written. I felt it’s an important case. Certainly, it’s something that I wanted to write about. It’s something that’s very important to me. [29:42] But it’s yet another attempt to apprehend these fugitives. And I’m hopeful that somebody, it might resonate with somebody. Somebody’s going to see something. And there’s so much disinformation and misinformation that’s out there in the media about this case. People are endlessly, all these armchair detectives, and I don’t say it in a deprecating way. Good for them. Work as hard as you can. But if you want to work this case from your armchair, great. but you should be going off accurate information because there’s a lot of bad information that’s out there on the internet. And if you want to help out, if you want to collect that $10 million reward, great, but you should be going off the most accurate factual information that’s available. Yeah. And you probably ought to go down to the deep seamy underbelly of Philadelphia or Boston or somewhere and get involved with a mob and then work your way up and make different cocaine deals and everything. And eventually you might be trusted enough that some might say, oh yeah, I’ve got those in this basement. I would suggest there’s better hobbies. [30:47] That could be hazardous to your health. I wouldn’t recommend it. Yes, it could. All right. Jeffrey Kelly, the book is 13 Perfect Tuesdays. Those are the paintings that were stolen that you’ll see on the podcast on the YouTube channel. The true story of the mob, murder, and the world’s largest art heist. Jeffrey, thanks so much for coming on to tell us about this. Thanks, Gary. Thanks for having me.

U.S. sales of certified organic products accelerated in 2025 reaching $76.6 billion with an annual growth rate of 6.8%, and House Ag Committee advanced a Republican-led farm bill 34-17 in a vote early Thursday.

Chris Markowski, the Watchdog on Wall Street, delves into the complexities of financial truths, political accountability, and the implications of U.S. military involvement. He expresses discontent with current leadership, questions the narratives surrounding war, and emphasizes the need for financial responsibility in funding military actions. Markowski also discusses the impact of tariffs on the economy and critiques government waste and inefficiency. The conversation highlights the importance of independent thinking and the need for transparency in political discourse.

In this weekend's episode, three segments from this past week's Washington Journal. First: a conversation with journalist Avi Mayer - founder of the Jerusalem Journal – discusses ongoing U-S and Israeli combat operations against Iran. Then: the first primaries of Campaign 2026 kicked off this week. We read the tea leaves with National Journal's "Hotline" Editor Kirk Bado. Plus: the video depositions of former President Bill Clinton and Hillary Clinton went public this week. We'll discuss the fallout and what's next in the Epstein files investigation with Ken Thomas of the Wall Street Journal. Learn more about your ad choices. Visit megaphone.fm/adchoices

Iranian president Masoud Pezeshkian is apologizing for attacks on its neighbours that are caught in the crossfire of Iran's war with the U-S and Israel. Despite his words, Iranian drones and missiles continue to strike Gulf countries. Pezeshkian is calling for a diplomatic solution to the crisis, but also rejects U.S. President Donald Trump's demand for Iran's unconditional surrender. Meanwhile, Israel says it has 'a well organized plan' to destabilize the Iranian regime and enable change. Also: The remains of troops killed in action have been returned to American soil. They were killed in Kuwait last weekend, on the second day of the war between the US and Israel, and Iran. And: Canadian businesses working on the weak spots in the country's food system. Trade vulnerability caused by the ongoing threat of U.S. tariffs is serving as fuel for businesses looking to have more of Canada's crops processed and sold at home. Plus: Tech startups growing in Newfoundland, Using 3D printing to build houses, Canada's strong start at the Paralympic Games, and more.

A Dubai resident has died amid more missile and drone attacks on gulf countries.It comes after Iran's president offered an apology to its neighbours, stating Iran would not strike them unless attacked first.The United Arab Emirates has come under attack from Iran, with debris from a drone falling onto the popular Dubai Marina and the international airport.US President Donald Trump has vowed to hit Iran very hard as the US and Israel launched big attacks across Iran. In Lebanon, there have been more strikes in the capital of Beirut, with tends of thousands of people displaced.----Meanwhile at home, hundreds of protesters gathered outside Sydney's Town Hall, calling for the monarch to return to power in Iran.Following military backing from the U-S and Israel targeting the Islamic regime in Iran, diaspora in Sydney took to the streets yesterday afternoon in support of US President Donald Trump.---A flood emergency warning remains current for the Katherine River.Authorities says major flooding is occurring at Katherine Bridge, where the river level peaked near 19.2-metres just before 11 o'clock last night.This is above the bridge deck and around the 2006 flood level.Further rainfall is possible for the next few days, which my cause renewed river level rises.---Meanwhile the weather bureau says south-east Queensland residents should be prepared for the risk of flooding today as a tropical low brings rain further south.The system made landfall near Innisfail on Friday and has moved into the interior before an expected path towards south-east Queensland.The bureau says the south-ease will see the heaviest of the rain today.---And South Korea says it will be going all out for the win in tonight's final group game against the Matildas in the Asian Cup.The South Koreans have a better goal difference than Australia, and only need a draw to top the group.Speaking through a translator, coach Shin Sang-Woo says it's an important match for his side. 

A week after U-S and Israeli strikes began on Iran, war is widening across the Middle East. Russia and Ukraine have carried out a prisoner exchange. Canada's prime minister has signed Australia up to a powerful global alliance of critical minerals producers during his visit. - アメリカとイスラエルによるイランへの攻撃開始から1週間。中東では戦闘がさらに広がっています。ロシアとウクライナが、捕虜の交換を行い、双方あわせて200人ずつが解放され、帰国しました。オーストラリアを訪れていたカナダのカーニー首相は5日、両国が重要鉱物に関する新たな協定に署名したと発表しました。1週間を振り返るニュースラップです。

The PM confirms three Australians were onboard a U-S sub that sank an Iranian vessel, Calls for action on petrol prices as the Iran conflict continues, Australia's flagbearers for the Winter Paralympics Opening Ceremony announced.

As work continues in Congress to piece together Farm Bill 2.0, there is also talk that lawmakers will soon resume work on farm labor reform.

The California Farmer Immigration Enforcement Survey is the first to assess the impact of the mass deportation agenda on California farms, and USDA announced the “One Farmer, One File” modernization plan, designed to put “farmers first” with technological improvements.

Georgia farmers, ranchers and foresters in eligible counties can begin applying for assistance for losses due to Hurricane Helene on March 16th, and USDA announced the “One Farmer, One File” modernization plan, designed to put “farmers first” with technological improvements.

Australian Foreign Minister Penny Wong says the world is worried about how Iran is reacting to U-S and Israeli strikes that killed its leader, Ayatollah Khamenei and targeted nuclear infrastructure. The Department of Foreign Affairs has sent the first of six crisis response teams to the Middle East, where thousands of Australians are caught up in a widening conflict that has seen strikes on Gulf states and enclaves that are heavily populated by expats and tourists. The far-north of the nation is on cyclone watch, with Cairns expect to bear the brunt of a tropical cyclone forecast to cross the Queensland coast tomorrow morning. News from today's live program (1-2pm). - オーストラリアのウォン外相は、米国とイスラエルによる攻撃にイランが今後どう反応するのか、世界は懸念していると述べました。中東地域で立ち往生しているオーストラリア人旅行者や現地滞在者の帰国をオーストラリア外務省が支援しており、今朝までにドバイから新たに230人以上が帰国しました。クイーンズランド州北部の沿岸部でサイクロンへの警戒が強まっており、ケアンズでも注意が呼び掛けられています。2026年３月５日放送。

Explosions have lit up the skies over Beirut as Israeli strikes continue to target Hezbollah positions in the Lebanese capital, widening the regional war that began after the U-S and Israeli strikes on Iran. Officials say more than 70 people have been killed in Lebanon, while tens of thousands of civilians have been forced to flee their homes.

Europe's industry body warns the U-S over punishing Spain for not sharing military bases, DFAT deploys teams to support Australians caught in the Middle East as the war widens, Arsenal defeat Brighton, getting closer to the English Premier League title.

U.S. Representative Dan Newhouse will spend his final term in office as Chair of the House Subcommittee on Forestry and Horticulture.

Subscribe here to Inside Call me Back ------- Please take 5 minutes to fill out Ark Media's LISTENER SURVEY ____ Was America dragged into the Iran war by Israel, as some commentators seem to suggest? An out-of-context statement by State Secretary Marco Rubio seems to give credence to the conspiratorial accusation that America is fighting Israel's war in Iran. Dan is joined by Nadav Eyal and Mark Dubowitz to discuss the truth behind these claims, the delicate politics behind the decision to launch the war, and how it all plays out in American public discourse. They also cover the possible succession of Iran's Supreme Leader, the campaign against Iran's nuclear and missile capabilities, and the broader strategy shaping the conflict—along with what outcomes Israel and the U.S. would ultimately consider a victory. In this episode: - Strike in Qom and the disruption of Iran's leadership succession - The possible rise of Mojtaba Khamenei - What the past 72 hours reveal about the military campaign - Why Qom matters inside the Islamic Republic - The emerging strategy to pressure the regime internally - The debate in the U.S. over how the war began - Iran's strategy for surviving the conflict - What Israel and the U.S. would consider a victory More Ark Media: Want to join Ark Media? Check out our careers page for new openings. Explore Israel Votes Listen to For Heaven's Sake Listen to What's Your Number? Watch Call me Back on YouTube Newsletters | Ark Media | Amit Segal | Nadav Eyal Instagram | Ark Media | Dan X | Dan Dan Senor & Saul Singer's book, The Genius of Israel Get in touch Credits: Ilan Benatar, Adaam James Levin-Areddy, Brittany Cohen, Ava Weiner, Martin Huergo, Mariangeles Burgos, and Patricio Spadavecchia, Yuval Semo

On this Salcedo Storm Podcast:Conservative Texas Attorney general, Ken Paxton.

The Daily Quiz - Music Today's Questions: Question 1: Who was the first Spice Girl to have a solo chart single? Question 2: Which band includes 'Anthony Kiedis'? Question 3: Which American rock band released the song 'Alabama Song'? Question 4: Who Released The 70's Album Entitled Trans-Europe Express Question 5: Which band includes 'Anni-Frid Lyngstad'? Question 6: Whose Real Name Is Roberta Anderson? Question 7: Which American musician released the song 'Mr. Tambourine Man'? Question 8: Which English rock band released the song 'Us and Them'? Question 9: Which American rhythm and blues girl group released the studio album 'Destiny Fulfilled'? This podcast is produced by Klassic Studios Learn more about your ad choices. Visit megaphone.fm/adchoices

Good afternoon, I'm _____ with today's episode of EZ News. Tai-Ex opening The Tai-Ex opened down 94-points this morning from yesterday's close, at 34,228 on turnover of 11.1-billion N-T. The market plunged more than 700-points on Tuesday following a sell-off sparked by sky-rocketing international crude oil prices due to the U-S and Israel's war with Iran. Economics minister dismisses power rationing concerns Economics Minister Kung Ming-hsin says the government has secured sufficient natural gas supplies for this month and there will "absolutely be no power rationing." The statement comes amid concerns that escalating hostilities in the Middle East could trigger rationing. According to Kung, gas shipments for early March have already passed through the Strait of Hormuz, while supplies for the second half of the month have been secured from "alternative (可供替代的) markets." Speaking at a legislative hearing, Kung told lawmakers that he hopes the conflict will be short-lived and the government use of coal-fired power generation will be "a last resort." CDC warning Japan-bound WBC fans of infectious disease risks And, The Centers for Disease Control is advising people heading to Tokyo to attend World Baseball Classic games to take precautions against measles, flu and infectious gastroenteritis. The C-D-C is citing elevated disease activities in Japan for the warning. According to C-D-C spokesman Lin Ming-cheng, travelers should be aware of a rise in measles cases in Japan, which have reached their highest cumulative (累計) total for the same period since 2020, while flu activity in Japan also remains at a high level. Lin is also warned travelers about infectious gastroenteritis in Japan, saying the country recorded an average of 8.02 cases per clinic in the eighth week of the year, the highest for the same period in five years. US to offer insurance, escort to tankers traveling through Strait of Hormuz President Donald Trump says the US will start offering insurance and navy escorts to tankers travelling through the Strait of Hormuz, as oil and gas prices spike due to the war in Iran. About 20% of the world's oil and roughly (大概) a third of its Liquefied Natural Gas passes through the narrow waterway off of Iran's coast. Toni Waterman has more Argentina Navy Trial for Sinking of Sub Four former Argentine navy officers are on trial for the 2017 sinking of the ARA San Juan, which killed 44 crew members. Prosecutors say the officers neglected duties and ignored serious maintenance problems. The submarine vanished (消失了) on Nov. 15, 2017, during a trip after a training exercise. The crew reported water entering batteries, then lost contact. Investigators say water reached a battery tank and triggered a short circuit and fire. The submarine then plunged and imploded (內爆) at extreme depth. Relatives of the deceased crew members argue that the current trial falls short of achieving true justice. That was the I.C.R.T. EZ News, I'm _____. -- Hosting provided by SoundOn

USTR Chief Ag Trade Negotiator Julie Callahan recently addressed rumors that the six-year review of USMCA this year could turn into separate bilateral talks with Canada and Mexico.

In episode 2015, Jack and Miles are joined by host of Pop Mystery Pod & Lady to Lady, Tess Barker, to discuss… How’s The Right Handling It All? They Aren’t Doing A Good Job Selling It, David Ellison Gets The Warner Bros. Toy He Wanted So Badly, Jim Carrey Sparks Clone Conspiracy Theories After Receiving French Award and more! Markwayne Mullin: "It's up to the Iraqi people -- I'm sorry, the Iranian people -- to choose their next leader" Markwayne Mullin: "He has the ability to come back and ask Congress to declare war on Iraq" Can you tell me where Iran killed thousands of Americans in America? Paramount to buy Warner Bros Discovery in US$110 billion deal as Netflix bows out of race Netflix Backs Out of Warner Bros. Bidding, Paramount Set to Win Netflix says it bailed on WBD because of money, not Donald Trump ‘Horrifying’: Hollywood blasts Trump’s role in studio sale Trump allies claim victory as the Ellisons expand their media empire ‘David Ellison Scares the S— Out of Me’: How Paramount Beat Out Netflix, Won Warner Bros. and Will Change Hollywood Forever ‘Last Week Tonight With John Oliver’: Could New Merger Mean Cancellation? Leaked audio: Warner Bros. Discovery CEO David Zaslav tells employees Paramount deal felt 'whiplash-y' WHAT REALLY HAPPENED to JIM CARREY: BIZARRE APPEARANCE SPARKS CLONE THEORY & CONCERN FOR HIS SAFETY César Awards Say Jim Carrey ‘Worked on His Speech in French for Months’ Amid Clone Conspiracy Theory: His ‘Visit Had Been Planned Since the Summer’ Controversy Over Jim Carrey's César Awards Appearance — Fans Insist It's Not Him Makeup artist comes forward as Jim Carrey impersonator after viral César Awards appearance Reclusive Jim Carrey Looks Unrecognizable During Rare Red Carpet Appearance — as Fans Suspect Actor Had 'Tweaks Done to His Face' LISTEN: Rollerdisco by Black Moth Super RainbowSee omnystudio.com/listener for privacy information.

The United States is once again considering military strikes to curb Iran's nuclear activities and missile program. China has emerged as a particularly important partner of Iran, serving as the country's largest trade partner and one of its few sources of consistent diplomatic backing. For Beijing, the stakes in the relationship extend beyond energy security, but also include great power competition with the US and China's broader strategic ambitions in the Middle East. There are reports that Iran is close to finalizing a deal to purchase supersonic anti-ship cruise missiles from China.  To unpack China-Iran relations and these recent dynamics, we are joined today by Jonathan Fulton. Jonathan is a nonresident senior fellow with the Middle East Programs at the Atlantic Council and an associate professor of political science at Zayed University in Abu Dhabi. His research focuses on China-GCC relations, China's Belt and Road Initiative, and Chinese foreign policy.  This episode was recorded on February 26, 2026.   Timestamps:  [00:00] Introduction   [01:33] China's Interests in Iran and Possible Reactions  [04:55] Challenges to Diversifying Oil Imports   [09:40] Using Oil Purchases as Leverage with the US  [10:59] Frictions in the China-Iran Relationship  [12:41] Iran in China's Middle East Strategy  [16:00] Iran–China 25-year Cooperation Program   [21:56] China-Russia Coordination in Iran Strategy  [25:54] Tehran's Points of Leverage with Beijing and Moscow  [29:14] Potential Disruptors to the China-Iran Relationship 

A kiwi living very near a Middle East military base that is critical to the US says he feels safe enough for the moment, but his family has an overland evacuation plan just in case. The US State Department has today issued a "depart now" warning to Americans living in more than a dozen locations in the Middle East. Just outside Abu Dhabi is a military base that hosts U-S troops. It's reportedly been a target of retaliatory strikes. Jordon Buchanan, his wife and two young children moved to Abu Dhabi about ten months ago. He spoke to Lisa Owen.

Good afternoon, I'm _____ with today's episode of EZ News. Tai-Ex opening The Tai-Ex opened marginally higher this morning from yesterday's close, at 35,106 on turnover of 8.9-billion N-T. The market lost ground on Monday, as investor sentiment took a dive on the back of concerns about global oil prices after the U-S and Israel attacked Iran over the weekend. MOEA talks boosting energy imports due to Middle East tensions The Ministry of Economic Affairs says it plans to increase energy imports in response to the current situation in the Middle East. According to the ministry, it will diversify import sources and strengthen energy supply security. The ministry says although the conflict is effecting some shipments oil and gas import coming through the Strait of Hormuz, short-term energy shipment arrivals remain unaffected and long-term contingency plans are in place to maintain a stable power supplies. The economic ministry's statements come as Premier Cho Jung-tai has been meeting with senior government officials for talks on the impact of higher international oil prices. Cho says C-P-C has been asked to ensure uninterrupted supplies, and Vice Premier Cheng Li-chiun will convene a series of price stabilization meetings. Man detained over Taipei Main Station bomb and death threats A man has been ordered detained following his arrest for allegedly sending threats in late February to set fires, plant bombs and kill people at Taipei Main Railway Station. Prosecutors say they secured court approval to detain the Changhua County resident - who police say admitted making the threats due to stress and being in a bad mood. The Taiwan Railway Corporation and the Taiwan High Speed Rail Corporation received separate threats on February 25 and 26, respectively. Police heightened security measures as the threats were made just ahead of the 228 Memorial Day long weekend holiday. The suspect was arrested following a raided (襲擊) his residence on the night of February 27 and taken into custody. He is facing charges of public intimidation. Trump: Iran operations to last 4 to 5 weeks or "far longer" US president Donald Trump says American operations against Iran could last four to five weeks, but warned the US had the capability to go "far longer than that". He promised Washington would "prevail" (戰勝) after news that a fourth US service member has died. Nick Harper reports from Washington. Cuba Detains Panama Citizens for Propaganda Cuba's government says it has detained 10 Panamanian citizens it accused of creating subversive propaganda in the latest international incident involving the Caribbean island. Monday's announcement comes just days after a fatal shooting involving a U.S. boat in Cuban waters. The government said that according to initial statements from those detained, the Panamanians were ordered to leave Cuba “once the objective was achieved,” and that they would receive between $1,000 and $1,500 each upon their return. Officials did not provide further details, including the nature of the alleged (被指控的，據稱) propaganda. That was the I.C.R.T. EZ News, I'm _____. -- Hosting provided by SoundOn

As the conflict in the Middle East escalates, what does it mean for Canada and the rest of the world? Ferry de Kerckhove, a senior fellow in the faculty of social sciences, and graduate school of public and international affairs at the University of Ottawa, joins Andrew Carter.

Political pressure is intensifying over the release of Jeffrey Epstein investigation files as Hillary Clinton appears before U-S leaders, and Democrats accuse the Justice Department of withholding key records. - ሂላሪ ክሊንተን ኣብ ቅድሚ መራሕቲ ዩናይትድ ስቴትስ ኣብ ዝቐረበትሉ፣ ብሰንኪ ምዝርጋሕ ሰነዳት ገበናዊ መርመራ ጀፍሪ ኤፕስቲን ፖለቲካዊ ጸቕጢ እናለዓለ ይኸይድ ኣሎ፣ ኣባላት ሰልፊ ዲሞክራት ድማ ክፍሊ ፍትሒ ኣሜሪካ ቁልፊ ዝኾኑ መዛግብቲ ሓቢእዎም ወይ ከየውጽኦም ተሪፉ ክብሉ ይኸሱ። እዚ ክትዕ፣ ኣብ ልዕሊ ሓያላት ሰባትን ናይ ቀደም ምሕዝነቶምን ምትእስሳራቶምን ዝግበር መርመራን ቆላሕታን ደጊሙ ኣበራቢርዎ'ሎ፣ እቲ ውድቀትን ውርደትን ድማ ካብ ዋሺንግተን ናብ ኤውሮጳን ዓበይቲ ግሎባዊ ትካላትን ሰጊሩ ይዝርጋሕ ኣሎ።

In this weekend's episode, three segments from this past week's Washington Journal. First: U-S- Iran nuclear talks continue –as the U-S continues its military buildup in the Persian Gulf. We talk about where things stand and the risks of an all-out war with Rose Kelanic of Defense Priorities & Danielle Pletka of the American Enterprise Institute. Then: This year's tax filing system is well underway. We speak with former IRS Taxpayer Advocate Nina Olsen about what you need to know -- including changes from the passage of the "Big Beautiful Bill." And finally: author & English professor Dan Chiasson discusses his new book on Sen. Bernie Sanders' early political life, titled "Bernie for Burlington." Learn more about your ad choices. Visit megaphone.fm/adchoices

On this Salcedo Storm Podcast:Robert Montoya is an investigative reporter for the Texas Scorecard.

Political pressure is intensifying over the release of Jeffrey Epstein investigation files as Hillary Clinton appears before U-S leaders, and Democrats accuse the Justice Department of withholding key records.

During the longest State of the Union address in history, United States President Donald Trump has declared the U-S is in its "golden era," with him at the helm. The almost two-hour long speech was filled with dramatic moments, as the President boasted about his successes and outlined his agenda for the coming year.

Specialty crop growers are paying close attention as the USDA announced up to $1 billion, and possibly more, in bridge assistance payments available for Specialty Crop growers.

Today on the @EricMetaxasShow, I react to Snowmageddon in New York City, break down the Huckabee Tucker conversation and the Israel controversy, and talk through the U S hockey win, a Trump security scare at Mar a Lago, and what Iran and rising Islamist extremism could mean for American safety. I also share an update on my Revolution book, America 250, and why I believe the Super Centennial moment matters right now. TIMESTAMPS 0:00 Intro 3:28 Huckabee Tucker Israel Controversy 12:29 Hockey Win And Patriotism 16:03 Trump Security Scare At Mar a Lago 18:28 Iran, Safety, And The West 27:01 Eric's Revolution Book Update

War Room Massie Drops Redacted Epstein Names On House Floor, U.S. F-22 Stealth Fighters Deployed to Southern Israel Amid Iran Tensions, PLUS Trump SOTU to Focus On Economy & Middle Class

War Room Massie Drops Redacted Epstein Names On House Floor, U.S. F-22 Stealth Fighters Deployed to Southern Israel Amid Iran Tensions, PLUS Trump SOTU to Focus On Economy & Middle Class

Good afternoon, I'm _____ with today's episode of EZ News. Tai-Ex opening The Tai-Ex opened up 247-points this morning from yesterday's close, at 34,020 on turnover of 14-billion N-T. The market closed higher on Monday, as the market reopened following its 11-day Lunar New Year holiday, but the gains were capped amid concerns over U-S tariff policies. Opposition calls for talks on format of Lai's legislative report Opposition parties are calling for cross-party negotiations over how President Lai Ching-te should deliver a national affairs report at the Legislative Yuan. The statement comes after Lai announced that he's willing to deliver (發表) a national affairs report at the Legislative Yuan "in accordance with constitutional procedures" following a meeting with the heads of the five branches of government at the Presidential Building. K-M-T and Taiwan People's Party lawmakers are seeking a question-and-answer format. Legislative Speaker Han Kuo-yu proposed a questions-first-answers-second format after Lai hosted the meeting - saying he invited the president to personally explain issues of concern to lawmakers. According to Han, Lai "gladly accepted" the invitation. Lunar New Year ER visits down as more clinics stay open, UCCs expand The Ministry of Health says visits to emergency rooms over the Lunar New Year period were down from last year as more clinics stayed open and urgent care centers expanded capacity. According to the ministry, data shows there were 232,397 visits to hospital ERs from February 14 to 21 - and that total was down by 22.5-per cent from the number of visits reported during the same holiday period of last year. Visits to the hospital ERs over the holiday period this year were closely watched after what happened last year - when ERs across Taiwan became so overcrowded that the Taiwan Society of Emergency Medicine called the situation "unprecedented (史無前例的)". The government spent 1.6-billion N-T in funding to encourage medical providers to stay open during this year's holiday. Trump threatens higher tariffs for countries who "play games" after Supreme Court ruled tariffs illegal US president Donald Trump has threatened (威逼) to put higher tariffs on countries that want to "play games" after a key U.S. Supreme Court tariff ruling. Kate Fisher report from Washington Canada Calls OpenAI Reps to Ottawa The Canadian government has called representatives of ChatGPT-maker OpenAI to Ottawa. The company says it considered but did not alert Canadian police about the activities of a person who months later committed one of the worst school shootings in the country's history. Canada's artificial intelligence minister says he expects the company's top safety representatives to explain its protocols and how it decides to forward cases to law enforcement when he meets with them today. OpenAI said last June that the company identified the account of Jesse Van Rootselaar via abuse detection efforts. The 18-year-old killed eight people in a remote (偏僻的) part of British Columbia this month and died from a self-inflicted gunshot wound. That was the I.C.R.T. EZ News, I'm _____. －－－－以下為 SoundOn 動態廣告－－－－ 新感覺夾心土司 多種口味隨心挑選 讓你隨時隨地都有好心情 甜蜜口感草莓夾心、顆粒層次花生夾心、濃郁滑順可可夾心 主廚監製鮪魚沙拉、精選原料金黃蛋沙拉 輕巧美味帶著走，迎接多變的每一天 7-Eleven多種口味販售中 https://sofm.pse.is/8rcv29 -- Hosting provided by SoundOn

When your corporate job feels "secure" until it suddenly isn't, real estate can become the Plan B that turns into your best move…  In this episode of the #DoorGrowShow, DoorGrow founder Jason Hull sits down with John Casmon (multifamily syndicator, host of Multifamily Insights, and co-creator of the Midwest Real Estate Networking Summit) to break down how corporate professionals can transition into multifamily investing without becoming a stressed-out landlord. They dive into how John went from corporate bankruptcies to building a multifamily portfolio, what passive investors actually need to know before putting money into a deal, and why trust + clear expectations matter just as much as the numbers.  Jason and John also unpack what this means for property managers: how to align with investor goals, why the best operators project calm control (even in chaos), where syndicators hang out, and how PMs can position themselves to win more multifamily doors.    You'll Learn (00:00) Transforming Property Management: An Introduction  (00:59) John Casmon's Entrepreneurial Journey  (02:56) Transitioning to Multifamily Investing  (04:33) Understanding Investor Types and Property Management  (05:48) The Role of Property Managers  (07:49) Investor Control vs. Trust in Management  (09:33) Challenges in Property Management  (11:17) Aligning Goals with Property Managers  (14:19) The Real Product of Property Management  (17:14) Managing Investor Expectations  (19:50) Syndication: A New Avenue for Property Managers  (23:44) Legal Considerations in Syndication  (26:41) Calmness in Chaos: The Key to Success  (31:40) Partnering with Syndications  (33:54 The Role of Property Management in Syndication  (38:29) Finding Syndicators and Building Relationships  (42:24) Understanding Passive Investment in Syndication  (47:45) Identifying Your Investment Goals  (51:54) Assessing Risk in Real Estate Investments  (55:15) Choosing the Right Market for Investment  (01:00:12) The Three C's of Raising Capital Quotables "The first C is confidence. Confidence comes from preparation." "The investment itself, we got to go out there and execute. But that investor psyche is a completely different game."  "It is not your job to hope. Your job is to analyze the information in front of you and make an informed decision." Resources DoorGrow and Scale Mastermind DoorGrow Academy DoorGrow on YouTube DoorGrowClub DoorGrowLive Transcript Jason Hull (00:01) All right, five, four, three, two, one. All right, I'm Jason Hull, the founder and CEO of DoorGrow, the world's leading and most comprehensive coaching and consulting firm for long-term residential property management entrepreneurs. And for over a decade and a half, we have brought innovative strategies and optimization to the property management industry. At DoorGrow, we are on a mission to transform property management business owners and their businesses.   We want to transform the industry, eliminate the BS, build awareness, change perception, expand the market and help the best property management entrepreneurs win. Now let's get into the show. So my guest today, I'm hanging out here with John Casman, a multifamily syndicator, host of the multifamily insights podcast and the co-creator of the Midwest real estate networking summit. And in today's episode, John's going to break down how corporate professionals can transition.   into multifamily investing, how to find the best markets, how to raise capital effectively, and what separates successful operators from everyone else. John, welcome to the DoorGrowth Show.   John Casmon (01:10) Yeah, Jason, thank you for having me. I'm really excited to be here. Love the intro, your intro, not my intro, ⁓ but excited to be here and share as much as we can on our journey to help all of your listeners reach their goals.   Jason Hull (01:22) Cool. So John, ⁓ it's great to have you. I would love for people to hear about your entrepreneurial journey. How did you get to where you are now? And then we can get into your business.   John Casmon (01:34) Well, the short answer is bankruptcy, right? I worked for a couple of different companies that went through bankruptcy and that really made me consider my other options. You know, I was at General Motors back in 2007, 2008, 2009 when we went through bankruptcy and I was there and I watched what that did to a lot of my peers. I one day in particular when we were going to have a lot of layoffs, I went to work as late as I could. But when I got there, I had a red message, a little red dial on your phone.   for anybody who's worked in corporate and remember voicemails. So I had a red dot on my phone, picked it up, pushed the play button and my heart skipped a beat because I thought maybe I was getting to the can, right? And it was actually a colleague of mine who sat kind of kitty corner in front of me and he had been let go. He, you know, was diabetic. He didn't know I was going to pay for his medication. He just was venting in his voicemail. And I just remember feeling empathy for him, but also   a sense of I just never wanted to be in that situation. So it made me really start to think about Plan B. Eventually I moved to Chicago, realized real estate was going to be that path and learned everything I could about investing. So it kind of took me down that pathway to say, you know what, I need a Plan B because no matter what you do, when you work in corporate America, you do not control your future. You know, there's politics, there's policy, there's a lot of different things involved that you do not control.   And sometimes it does just come down to someone not liking you for whatever reason, or they think you're a threat. And I didn't want to spend the rest of my career navigating those issues. So I figured I had to take more into my own hands.   Jason Hull (03:16) got it. And so you start taking things in your own hands and what was the result?   John Casmon (03:20) Yes. So we landed on multifamily investing, started with small multifamily. My first investment was a two unit building. We house hacked it, which is a common popular phrase now. But back then it wasn't quite as common. But we lived upstairs. We rented out the first floor unit and it worked great. You know, it worked so great that we went to refinance and we had created enough equity in that first investment to pull out a six figure line of credit and go out and buy another property. So.   Jason Hull (03:45) Nice.   John Casmon (03:47) That really got the ball rolling. bought a three unit building, we bought an eight unit building, and at this time I'm still working in advertising, still working in corporate America, and I enjoyed what I was doing, and I just had my second child, but the agency I was working for also went through bankruptcy right at this time. We had expanded, we were growing, and we had kind of combined with a few other agencies and kind of became this little conglomerate, and it just eroded just as quickly as it grew.   I remember again, just sitting there and I've got some real estate. I've got a little bit of cashflow, but not enough to pay all my bills. New baby. And I just realized this real estate thing is working, but the exact strategy I'm employing doesn't allow me to insulate myself from these economic changes and shifts. So I had to change my strategy and that led me to syndication. Since then, we've acquired over $150 million worth of apartments.   We've partnered with busy professionals to buy these properties and give them some passive income. And that's what we've been doing ever since.   Jason Hull (04:50) Got it. So your area of genius really is helping these people that were similar to you, they're in the corporate environment transition into being an investor in real estate.   John Casmon (05:01) Yeah, exactly. And I would say too, it doesn't have to be you're going to quit your job and do this full time. And in fact, most people don't, you know, but most people do want a little bit more control over their life. You want a little bit more flexibility. You want to earn and start building up, you know, your net worth. You want to have a little bit more liquidity. You have to look at your investments to say, what should you be doing? I think most people know that their 401k, their, you know, company issued life insurance.   probably not enough to really get you on the fast track to retirement. So what else could you do? Certainly you can invest in the stock market. Lots of folks do that. But real estate is a proven vehicle. The challenge is, I don't know anyone who really wants to be a landlord, right? ⁓ Certainly you want the benefits of real estate investing, but very few of us want to get those 2 a.m. phone calls. So the shortcut there is, ⁓ hire a property manager. Great solution. But now you have to be able to manage   property managers, right, which is this whole other business. And if you don't have enough scale, then it's hard to get that person really focused on your business. So we offer an alternative, right? You get all the benefits of real estate investing, all the ownership perks without any of the headaches of being the landlord yourself. So it really is a great marriage of being in real estate without having to do the heavy lifting yourself.   Jason Hull (06:15) Okay.   Okay, so ⁓ the target audience of this show are property managers. So if they're not gonna use property managers, then what's the alternative? How does this work?   John Casmon (06:29) Well,   first of all, what we do is not always for that individual. So I think that's the key, right? You've got to understand who you are from a psychological standpoint. So when it comes to investors, there's two types of investors. One wants control, right? They're not willing to be passive. And some people think they want to be passive until they're in a passive situation and then they're calling and they want to know why you did this and why you did that and how come you did do that. That's not a passive investor. And that's fun.   Jason Hull (06:45) Yeah.   Yeah, they're anxious. Yeah. Yeah.   John Casmon (06:58) And   if that's you, you should be active, right? And you should work with a property manager, but you also want to work with the property manager who is going to be right for you, right? Because sometimes that is not how they operate. So you want to understand that. And that's a process to understand who you are as an investor, what kind of investment strategy fits you and what's going to be right there. When it comes to property managers, though, I think there are a couple of things. And as a matter of fact, we just left out of meeting with   property management company yesterday. They have 2000 units. We talked about some other services that we offer. And one of things that stood out to me was just understanding some of the challenges that property managers face. And one of them is property managers are really in a position to think like everyone. They're supposed to think like an investor. They're supposed to understand maintenance and kind of the construction arm enough to understand what needs to happen at a property. But they are really little CEOs, right? Because for   Our stuff, the large apartment stuff, those are typically million dollar annual revenue businesses. And this person is in charge of that asset of that business. They are making the day to day decisions. They are the face for the residents, aka the customers of that business. They are the face and their experience with that individual is how they view that business. So it really is an important role. And if you're working with property managers, it's really important to understand how to find the right people.   to connect with them and have them represent your business, your brand, company in the right light.   Jason Hull (08:30) So now you left an open loop that I want to close. So you said there's two types of investors, those that want control and maybe should go find a property manager, you said. And then what's the other type?   John Casmon (08:34) Yeah.   The other type is those who don't want control and they trust someone else to handle that. And for them, there are a couple of different ways of investing. One is investing passively with a group like ours. The other is turnkey investing where again, you hire a property manager, but you really entrust them to manage the property. The only thing I would say for either one of those groups, myself included, is you want to trust but verify. Okay. You've got to do a lot of your due diligence upfront. You want to understand how they operate. You want to talk to   some of their other clients, some of their other investors, because you need to get a really good sense of what to expect. And a lot of people are great at selling themselves upfront, right? I can tell you everything you want to hear upfront. You want to know what is it like once you sign the paperwork? How often are we going to talk? How frequently am I going to get updates? And at what point am I able to weigh in and make decisions? Because if, if you are someone who wants to be more active or be heard, or you've got thoughts and opinions,   Jason Hull (09:18) yeah.   John Casmon (09:35) You want to make sure you have a voice in your investment. Otherwise you may get really disappointed or you may bring on someone who has a different perspective of what that relationship looks like and that never is going to work out.   Jason Hull (09:47) Yeah, there's a big challenge in the industry and that's that most property management companies suck. so most investors that have dealt with property management to some degree are they have some scar tissue, they've been burned a little bit. They've a lot of property managers that started their businesses that come to me for help to grow their business. They started because they were investor and they couldn't find anyone else to manage the property good enough. And that's why they started their business, but it can be a difficult business to run. so none of them start their business saying, I want to suck.   But that's kind of the default unless they get some really good support or figure some things out through a lot of trial and error. And so that's where DoorGrow comes in. We help them with that. But one of the things I coach my clients on a lot is that they need to shift into being daddy over these rental properties. They need to like tell the owner, hey, you need to trust me. And they need to be able to have a really effective business so that they can lean into that trust.   because a lot of people are anxious. They'll come to them with concerns, but generally if a property manager is good, they're much better at this investing stuff than most investors. And they're much better at coordinating maintenance. They're much better at handling leasing. And so when an owner tries to micromanage a property manager, it kind of doesn't make sense to hire somebody to manage your asset just so you can manage them to do the job. And so I think the secret is finding a really good property manager that you can   let go of control because you can trust them. And but yes, you need to verify that they can do the job that you need them to do. And so a good property manager will take ownership of it and they'll take control and they will, they'll display a lot of certainty and confidence in how they communicate and they won't allow you to micromanage them is what I've seen. So.   John Casmon (11:37) Yeah, Jason, and I'll add to it. There's a two way street there. And I think it's easy for people to say, ⁓ most property managers suck or they're not good or whatever. And listen, there's certainly a lot of challenges there. A lot of folks who are not living up to par to the standards. But I will go back to this. We ask property managers to do the work of generally like a CEO. Right. I mean, again, they're managing million dollar businesses in many cases, yet they don't have that training. They don't have that experience. They don't have the ability to navigate.   all of these various things. So part of what owners and investors need to also understand is that you play the role of asset manager. And that means giving clear direction of what success looks like so that that property manager has a framework to make decisions. It's not to micromanage those decisions, but to help them understand how their decisions impact the greater good. And part of that is like, again, just sitting down with annual goals. What are revenue goals? What are our goals on?   Occupancy, what are our goals on in a lot? And this may seem simple, but I promise you a lot of folks don't do this. And if you don't do that, then that property manager is going to default to, for instance, I'll give you a great example. I've got a property manager. She's awesome rock star. But she always gets nervous when occupancy is not at like 96 or 97 percent of this property. So she is, you she starts apologizing profusely and all I did this or done that and like.   Jason Hull (12:58) Yeah.   John Casmon (13:04) Occupancy is one of our KPIs for sure. It's important, but that is not the KPI. I am focused on my net operating income. And if we're going to push rents, the impact of that is you're going to have higher vacancy and she is not comfortable with that. And that's probably because she's used to working with owners who want that thing fully rented and they are comfortable having 100 % occupancy.   Jason Hull (13:13) Yeah.   Hmm.   Yeah.   John Casmon (13:33) if they're leaving 50 bucks, 75 bucks, whatever it is of rent on the table. And that's the part where you've got to really align with your vision versus their vision, because what they have in the back of their mind may not completely align with what you have. Or they have residents in their face who are coming into the office. They want something fixed. They want it done quickly. They want it done right. They want it done yesterday.   Jason Hull (13:49) Right.   .   John Casmon (13:59) So they've got that pressure of this person in their face. So they may go out there and spend the money or authorize the money to get spent. And maybe they're not picking the most cost effective measure. So you have that. And I'll give you one third one. A lot of times when you run into the flip side of that is maybe occupancy is low. They say, hey, we need to increase our marketing spend, right? We got to increase our marketing budget. know, ox is down to 88 or 90%. We got to spend more money. And we're not necessarily.   really zeroing in on what the specific issue or challenge is at that property. So for an owner, your job as an asset manager is to partner with them and to help them see what the options are, help them work through with some of those challenges and solutions are and partner with them to find success. It's not to micromanage them and tell them what to do, but it's really to understand the situation better and give them that perspective.   Jason Hull (14:49) Yeah, that makes a lot of sense. think, you know, one of the things I've seen is that I've noticed a lot of property managers, they make the mistake of thinking that the goal or the product that people want to buy from them is property management. But investors don't wake up in the morning and go, man, I'm so excited to get property management today. The thing that they want. And so the way I describe it to them as they say, property management is like the flight to Hawaii. It's not Hawaii.   and you're trying to sell the flight. That's not the exciting part. You need to figure out what the investor wants, what their goal is. Where do they want to go? What's Hawaii for them, right? What's paradise? And then how do we optimize for that? And how do we help them create a path for that? Because the actual product that a property manager is selling is not what they do. It's not property management. The actual product is them. It's them and their values and their belief system and how they create trust and the team they build and the system and mechanism they build around them.   That's the actual product the property manager is selling. so a lot of property managers make that mistake. They sit there and talk to you about maintenance coordination and leasing and inspections. And meanwhile, you're just wondering as an investor, can I even trust this person? Like do our values align? Yeah. So I don't know what your thoughts are on that, but.   John Casmon (16:11) I think you're spot on, right? Because, I mean, ultimately, as an investor, you are only as good as the team you can build. And that property manager is in charge of the day-to-day aspects of the business. especially when you, you know, I've heard horror stories of folks who have done like turnkey investing, right? Where the property manager, someone owns it, they buy it, they fix it up, and then they rent it back to...   an investor. And I've heard horror stories where that property was not being well managed. And that's the fear. If you're not in that marketing, you can't come and see it. So if you got an out of town investor, you really are trusting that property manager. So that is the most important thing, right? Everything else are tactical, daily situational things that can change. But it comes down to do I have the right people, people that I can trust, people who are going to make the right decision based on the information they have.   because they may not know what I know or maybe something shifted and changed where they would have made a different decision. We can't, you know, ache on that. It really comes down to are they doing their best? Are they making good decisions? If they're not making good decisions, is it because they didn't have the correct information, which again, could fall back on you as the investor to say, hey, are they aware of what your goals are? Are they aware of maybe this situation, these tools, these resources, whatever it is? And that's on you to sit and collaborate.   But trust is absolutely paramount because at end of the day, the thing that I think most of us are concerned with is who we partner with. And there's a great book I'm reading right now. And it gets into decision making and the fear of decision making for most of us and why deals stall. Why didn't you hire somebody? Why didn't you, you know, go with the vendor or go with the contractor or with the company? And the biggest thing is we are scared of making the wrong choice. All of us in decision and no action.   Jason Hull (17:43) Absolutely.   John Casmon (18:04) is better than the wrong action for many people because they once they take action. Well, now they're blaming themselves because you didn't pick the right person. Why did you hire that guy? You should have like now this starts to go on in their head versus doing nothing. Well, at least it's you know, it's not going to get worse, you know, it will in lot of cases get worse. So for a lot of people, that is the scariest thing. So if you can take that fear off the table as far as being the right person or being someone who is trustworthy.   Jason Hull (18:07) Right, yeah.   John Casmon (18:32) everything else gets easier. So if you can do that, that's, you know, the best thing you can do as an investor or as a property manager.   Jason Hull (18:38) Yeah, I agree. think one of things that I talk about a lot is that clarity has to come before action because if you don't have clarity and you start taking a bunch of action, doing stuff, every action you take is a little bit wrong. Sometimes it's a lot wrong. so, yeah, we need to get that clarity first before we start ⁓ making moves. And you talked about, I love the example of your property manager that is trying to   optimize maybe for the wrong thing. They're like, want to optimize to the, making sure their vacancy is super low. But that might not be the goal. That's not the primary goal. The goal is money, you know, and there's a really good book is by Elihu Goldratt. It's a good book for operations people, but it's called The Goal. And spoiler alert, the guy's trying to figure out the goal through this whole book, the story and it's money. That's the secret. The goal is the of the business, should be making making money.   And what happens in this book is that people are over optimizing individual pieces in this flow at this warehouse. And it's actually not helping to make money. It's causing more constraint. And so if we over optimize at one stage, it actually creates waste, bloat, inventory, additional work for the next stage. And so sometimes the best thing certain departments can do is slow down and do less in order to get the outcome to be maximized outcome.   And there's some really great examples in that that I think are really powerful. But I think the if you're optimizing for the wrong thing, then you're not making it effective. So you want to make sure you're optimizing for the right thing. Otherwise. ensues. You get mad at somebody, but nobody understood what the goal was. And so I think, yeah, getting a greed upon set of criteria of what what the outcome is and asking the property manager, can you help me achieve this?   And they know, they know if they know what the problem is, usually they can, they know how to help you get whatever goal that you have. And they know whether your goal is probably realistic or not, because they've helped probably a lot of people do this similarly. And so, but yeah, I think it's very important. Make sure you know, where's Hawaii and maybe property management is the vehicle. Now you had mentioned like, I'm really curious about this idea of, you know, maybe creating syndications.   Some property managers are now starting to think, maybe I should create a syndication. What's your criteria for, what's a good syndication and what are some of the, I'd be really curious to get into if some of the property managers listening were wanting to do kind of a little bit of what you do, how they might be able to get started in that. Like what are the beginning steps to make sure they don't make the mistakes you probably already figured out in the beginning?   John Casmon (21:27) Well, I think the first thing is, you really want to get into it? Right. Because for a lot of people, you got to understand it's a different business. Now you're not talking about real estate investing. You're not talking about property management. You're really talking more about, you know, investment management. You're talking about bringing on private investors who are looking for a return. That is communication skills. That's building up a network and a database of   Jason Hull (21:35) Mm-hmm.   Right, returns.   John Casmon (21:54) prospective investors, it's understanding the return projections that they're looking for. And it's really kind of managing the investor expectations, not necessarily the investment. And to give you a great example here, I had a deal where the investment went great, but it was slightly lower than what we initially projected. And I had an investor who was upset.   Jason Hull (22:07) Yeah.   Yeah.   John Casmon (22:23) about that. And we had communicated all throughout the entire process where things sat and he wasn't too upset, but he still made it a point to let me know, hey, well, this is less than what you initially thought. And that's challenging because the market shifts, right? Anybody who's bought properties in 2022 and beyond knows the market has shifted drastically over the last three or four years. So those projections made in a 2021-22 environment   Have a hard time standing up in a 25 26 environment We still make good money on that deals double-digit returns for investors ⁓ But you know there was that that was that feedback I got from one of the investors conversely We just exited deal a couple months ago, and we completely exceeded our return projections You know we delivered on a almost a 2.7 equity multiple Hit all you know mid 20s on the IRR completely unheard of stuff in this environment   And I have one investor call me and say, hey, John, I just checked my account. Is this right? And I'm like, yeah, it's it's right, man. He's like, my gosh, you guys killed it, man. my. Like, this is amazing. And it's great to hear. But again, that is separate from the investment. Right. Happy to manage the investor expectations and concerns. But that was an up and down investment where we had, you know, a moment where we actually had to put some of our general partner capital into the deal to keep it going.   Jason Hull (23:27) Yeah.   Yeah.   John Casmon (23:48) We have floating rate debt. had to refinance out of that. And we had to kind of rush to do that before rates started to go crazy. We had moments where our construction or renovation costs were much higher than we anticipated. So there are a lot of things that we had to navigate. And I think what happens for a lot of operators, a lot of people who get into syndication, they know the real estate and want to do the real estate, but they do not understand the perspective of the investor. And when you don't communicate to investors on a frequent basis and a clear, transparent nature,   Jason Hull (24:19) Yeah. Yeah.   John Casmon (24:19) They fill in the blanks and   the first concern every investor has and they won't say it. Most of time they don't say it, but I promise you they're thinking it after they make that investment. my gosh, did I make a mistake? Am I going to lose money? Is this person going to run off? Is this going to be some sort of fraudulent thing? Is this deal going to fail? These are all that we're wired like that. This is caveman stuff, right? We're wired to protect ourselves.   Jason Hull (24:36) Hmm.   Right.   John Casmon (24:45) And when you make an investment, and by the way, our investments are typically $50,000 and up, right? So these are not small investments. So when you make that investment, people start to second guess that decision. So my job when it comes to this side of the business is to keep them grounded that, hey, you've done your research, you've made an informed decision, you've picked a good partner, we've done this before. ⁓   Jason Hull (24:50) Yeah. Right.   John Casmon (25:13) And it's really to make sure that they feel comfortable with that decision. It has nothing to do with the investment, right? The investment itself, we got to go out there and execute. But that investor psyche is a completely different game. So first thing I would tell any of your property managers when they get into this business is understand, do you actually like people? Do you want to manage investors? Are you comfortable managing people's money? ⁓ And then beyond that, you have to do it the legal way. There are a lot of regulations around accepting capital from other people.   Jason Hull (25:31) you   John Casmon (25:42) So you can do it as a joint venture. The more common way of doing it, the more accepted way of doing this is by doing a formal syndication, which requires you to file SEC documentations. ⁓ know, there's regulation D and regulation A and there's some couple others, but typically it's going to be reg D 506 B or 506 C filing, which basically is the the structure that allows you to offer ⁓ passive investment opportunity or a security to investors. So again, for some people,   It's overwhelming. they're like, nope, never mind. But for some people, they love it. They want to get into it and they can learn more about that process.   Jason Hull (26:19) Got it. Yeah. I think I love your idea that it's more about managing expectations rather than the investments. And I think, I think that's good advice for all the property managers listing. This is something we spend a lot of time coaching clients on because they think their job is to manage properties. But really, if they're not strong in managing expectations and managing the relationship, it's 10 times to 100 times harder to manage the properties.   their operational costs go through the roof because owners are getting anxious. They're asking more questions. They're getting all these interruptions and calls, tenants, owners constantly. And if they had just managed the relationship and expectations and set strong boundaries at the outset, everybody would feel calmer. And I think really for business owners, I think the thing that really stood out to me that I've been focused on, and this is I've done some personal coaching and this is just nervous system regulation.   If you can, and John, seem like you're pretty chill and pretty calm and I'm sure the investor feel safe with you, which is why you've had success. If you are a person that is anxious and you're running around like a chicken with your head cut off, you're going to have, you're going to struggle in leading anybody, especially in relationships to your spouse and like everybody else. so having a calm, regulated nervous system allows your investors.   to entrain to your nervous system and to feel safer and to calm down. And that's not something you can pretend or you can just fake. You have to be that and they can sense and they can feel that it'll come across in your tone and in your body language and how you communicate. But if you can make sure that you're in that space and that you're able to regulate your own system, you're able to stay calm when other people are coming at you.   and other people are angry and other people are emotionally heightened. And you recognize this isn't really you. It's just that's them. And you can maintain that calm. You will be able to create a lot more safety. And that's really what people want to buy. Most people out there, their primary basic need is safety and security. Most people. That's why they aren't entrepreneurs. That's why they don't go start jobs. That's why they aren't like you and me. And if you're a property management business owner listening to this,   Most people are not like you. They want safety and security. That's why they get a property manager. They want peace of mind. And so, and I'm sure investors in a syndication, they also want some peace of mind because this is a big chunk of change.   John Casmon (28:55) They do. And I will say to most of the property managers I come across thrive in chaos. Right. They're used to stuff getting thrown at them. Right. And when you talk to them and get to know them, you learn very quickly. They like it. They do. They like the fact that they don't know what the day is going to bring. It could be a. Yeah, yeah. Could be a tenant coming with some crazy issue. It could be something from it's never boring and they thrive in it. However.   Jason Hull (29:00) Yeah.   Yeah.   They like the variety and unique challenges that property management brings, for sure.   It's never boring.   John Casmon (29:25) What happens then if you if they're going to look to work with investors and particularly raise capital and kind of do their own syndications, they have to understand that while they may thrive in chaos and uncertainty, most other people want organization. You want everything you said right. You want to have the calmness. You are looking for a captain to steer the ship. And for that part of the personality, they're going to have to tap into a different side of it to demonstrate how they handle chaos.   Jason Hull (29:37) Hmm.   Yeah.   Yeah.   John Casmon (29:54) not that they are chaotic. And I think what happens a lot of times when you're working with property managers is that they don't project that level of control. It just feels like they're reacting. So part of it is that, and they're really, really good ones. The ones who make it to that next level who are the regional managers and get those promotions, well, that's what they do. They manage the chaos and they manage up. They do a great job of telling the owners,   Jason Hull (30:06) Yeah.   Mm.   John Casmon (30:23) the leadership, whoever they need to talk to, they're telling them, hey, here's how here's our process. Here's how we're managing the situation. Here's what's going on. Here's what we're into. Hey, we had a water main burst here. Here's we bought. call three companies. We've got three quotes, but it's calm, right? It can be the worst. I'll give you a real example, right? At a fire, one of my properties and I was going to meet a property manager and I just happened to have a meeting with her that day at the property. She called me.   I was literally about to get in the car. She called me and said, Hey, I just want to let you know we've got a fire going on at the property. I'm not sure if you still want to meet. You're happy to come. We already have, you know, the fire department's here. They're they're putting the fire out right now. We already have another company that's coming in. They're going to walk through the damages once this is kind of settled. And I've already talked to the residents. Residents are good. We've got them hotels for the evening. We've checked with insurance. This is covered in your policy. So they're good to go. So you're happy to come down and talk and all of that if you want to.   Or we can let things settle down and maybe we can meet next week. This is a fire, right? This is like a scary situation. She called me.   Jason Hull (31:26) Right. A literal fire. Yeah. And there's plenty of fires   in managing properties. The literal ones.   John Casmon (31:33) Her calmness, she was so calm. Not only was   she calm, she had handled 90 % of it, right? It was the stuff you could handle in the moment. She handled it. So was like, hey, I don't think it makes sense for me to because I'm probably just going to add more anxiety to the situation at this point, right? It seems like you've got it under control. Why don't we let things settle, literally let the dust settle? And then once it's there, I'll come down. We can assess the damages, figure out what else needs to happen, what other next steps need to take place, right?   Jason Hull (31:41) Yeah? huh.   question. Yeah.   John Casmon (32:03) but had it handled like a rock star. Now, a lot of other folks would have saw the flames, called immediately, my God, there's a fire. ⁓ my God, what are we gonna do? So now you freaking out, everyone's freaking out, no one's controlling the situation, right? So now everyone's mind is just spinning and going. it does really take, kind of go back to where we started the conversation, that mindset of someone who was the boss, who was leading.   Jason Hull (32:05) Yeah, I love that.   Yeah. Freaking out. Yeah.   Hmm. Yeah.   John Casmon (32:32) who is going to take charge, even though it's not their property, they're going to take charge. Here's what needs to happen next. Maybe you have an emergency response plan already put in place, but you have these things already scheduled and ready to go. So when they happen, you're not shocked. You're not surprised. You're not asking questions that maybe you should have figured out upfront. And that's what a great property manager does. And if you convey that to owners, you're going to stand out above and beyond your competition because most people cannot convey that level of control, the level of   planning and the level of expertise that it takes to truly and effectively manage properties from the front, being proactive as opposed to just reacting to whatever the issue of the day is.   Jason Hull (33:13) Got it, okay. So ⁓ I'm reading, I just read, well, I didn't just read. I read in the past a really great book called Extreme Ownership. Really good book. Yeah, phenomenal book. ⁓ I'm going through their newer book, which I think is even better, called The Dichotomy of Leadership. leadership is what we're talking about right now, is that that,   John Casmon (33:23) Yeah, I think I got it like right here. It is right there.   Absolutely.   Jason Hull (33:38) creates a huge impact and there's a lot of misunderstandings of what leadership is, like it's control or it's being aggressive or, but yeah, it's really that calm presence of letting people know I've got it. Like we can take care of this. We've got a plan and staying regulated and calm. So I love that. ⁓ have a, so another question I have is how can the property managers listen to this? How could they maybe target or partner   with, if possible, syndications like you, like people that are doing what you're doing. Is there a chance that they could be a resource or do most syndications just in-house and do, they are a property management business?   John Casmon (34:19) No, no, most ⁓ most that I know work with third party manager companies. So I would say first and foremost, if you and syndications, I mean, it sounds like a big, huge, fancy word. But I mean, honestly, anytime you work with passive investors is technically a syndication. So it really comes down to figuring out who is looking for third party management and whether or not it's technically a syndication or not is really irrelevant. You want someone who is going to be managing or owning the property.   Jason Hull (34:24) Okay.   Yeah.   John Casmon (34:49) They want third party, but you have to understand their plan, going back to understanding the goals, right? Most syndications are looking to sell in a three to seven year timeframe, typically five to seven years. Most buy and hold owners have not decided or have not identified their exit strategy. So that's probably the biggest difference is when you have, let's just call it an individual investor or maybe it's a   Jason Hull (35:01) Okay.   Right.   John Casmon (35:17) a family or whatever that's buying and they want a third party manager, they don't know the exit. They haven't predetermined that they're going to sell in five years. So they are buying and holding it. And that goes back to the the I think the separation of understanding the objective, because for that person, having a full property is great. It means they're maximizing the revenue potential today. When you are syndicating.   most syndicators already assume 5 % vacancy. That's that's in everyone's underwriting. So you being at 100, they won't even give you credit banks don't even give you credit for it. So all of these things are already assumed. So for us to be above that is actually a miss, because it means we're not being as aggressive on the rent. So just understanding the mindset of a syndicator, which is they are looking to sell typically they're looking to double their money over a five or six year period. So how can you create value?   And that's something most property managers don't fully understand. But I would sit and I would talk to that syndicator. And if you want to be a syndicator or partners, not just be a third party vendor, but you actually want a partner, which we have seen a lot of folks look to do. You want to figure out how you can bring value to the table, because now we are aligning your interest with that syndicators interest. And now you've got a great partnership.   because every syndicator is going to need property management and they're going to need construction management to drive value. So if they can bring those people in as partners, that's a great opportunity for you. And if you're a property manager, you may have phenomenal relationships. You may already have contractor or the vendor partners that you trust in that marketplace. And if you could then take that and get a slice of the equity, that makes you very valuable for both sides.   Jason Hull (37:08) Do syndications, do they also need investors in capital or do most of them have that, are they really good at that? Okay.   John Casmon (37:15) Absolutely. Yeah. Yeah. Yeah.   mean, I mean, syndication at its core really just comes down to the need of capital. If someone had the capital themselves, they would probably just buy it directly and not go through the process of syndication. Because the syndication is literally just raising the money from passive investors. And in that scenario, again, being able to manage that, manage the communication, ⁓ that's really what a syndication truly is.   Jason Hull (37:42) So a really good property management partner could bring property management, some of the construction elements and investors and capital to the table. So it could be a nice little.   John Casmon (37:51) That would be amazing.   I'll be honest, man. That's because I don't want your listeners sitting here like, oh, I don't have one of those. I don't know if I've ever met one that had all of those. If you do have all of them, yes, you should consider syndicating yourself because you got all the pieces to the puzzle. Typically, what happens is a property manager has the property managers. I'll give you a great example. I got a 54 unit down in North Carolina. OK, so I came in as a key principal. I've got a.   Jason Hull (38:03) Okay.   Okay.   John Casmon (38:20) to my coaching clients. It's his property that he found. He asked me to come help him with the loan, which I did. One of the members, one of the partners is the property manager. So that's kind of their role to the table is they're managing the property. That's what they kind of came on. They had a couple of relationships, but their main role is the asset and property management side of it. So that's a great way to come to the table. But. Just like anything else in business.   Jason Hull (38:33) Mm-hmm.   John Casmon (38:49) It's very hard to find someone who checks every single box. I mean, that's like finding the marketer who's a CMO, who's also the CFO, who's also the COO, who's also the chief of human resource. very like no one, people don't really have like top notch excellent skills at every single one of those, right? Like you might be great at business, great at sales, great at marketing. You're probably terrible at finance, right? Like you just, you just forget to do your expense report type person, right? So it's hard to find someone who's   checks all those boxes. And I think typically when comes to property management, you want someone who's great with people, can resolve issues, but also has to be somewhat, you know, sufficient when it comes to the numbers, tracking all the data, tracking all the, you know, the rent roll, the leases, the income and expense statements, things like that. So usually they're not going to do every single box. But again, if you can find someone or that's where partnerships make sense.   Jason Hull (39:24) Mm-hmm.   John Casmon (39:43) If you've got that awesome. And again, I'm not saying a company doesn't have that. I'm just saying a single individual doesn't, which is why it's great to partner. If you can find someone who maybe brings a set of skills that you don't have, whether they're joining you in your property management business or they're partnering up where you're bringing your property management skills to the table with their investing or their networking skills, that makes for a good partnership.   Jason Hull (39:43) Mm-hmm.   Yeah, I got it. Well, we've got several clients, you know, all over the U S that are really good at property management. They're really good at handling the maintenance stuff and they obviously have a pool of investors as clients and, and, know, and they know that they can't do everything. So we coach them in making sure that they would do time studies. They figure out which, what their purpose is. We start to align them towards more fulfillment, more freedom, more contribution and more support in their business.   John Casmon (40:32) Yeah.   Jason Hull (40:38) And they start to build the right team. So they're getting operators, they're getting BDMs, they're getting the things they're not like strong in. And so we just make healthier businesses. So for those of maybe my clients listening that have healthy property management companies. And, but they don't want to do syndication. They're just like, man, that's a whole nother business. If I stay in my lane, I can grow that faster. How do they find syndicates? Like, how do they find people like you? Cause you've got a lot of properties connected to you.   and they would probably love to chat with somebody like you. Where do you syndicate people hang out? What's the title? Who runs a syndicate? What are they called? Do they have a specific title?   John Casmon (41:15) You   Yeah.   Yeah, great. Great question. Multifamily syndicator is is kind of the name just syndicator. We're all over. So I've got a podcast called Multifamily Insights. I interview like minded individuals. I've been doing that for a long time. We've done our seven hundred and seventy plus episode. So lots of people, lots of syndicators there. Definitely conferences. So if you look up any multifamily conference in your city.   Jason Hull (41:25) Okay.   Nice.   Okay.   John Casmon (41:46) meetups, lot of meetups in different cities as well. Those are great places to find syndicators. I think the biggest thing though is this.   Figure out who your avatar is. Because while we're talking about syndicators, ultimately, if you want to scale your property management business, I presume you're trying to scale with folks who are looking for third party management and the best option for that. OK, and let me back up. had one of the guests out of a podcast some years back, ⁓ Ashley Wilson. Love Ashley. As you said, something really changed when I thought about the business.   And she said the best way to find any vendor, any vendor is to figure out who relies on that vendor next and ask them for referral. So if you think about it, if you want a great drywall person, ask a painter. A painter is going to know who's great at drywall because they're going to know who makes their job easy and they can come in and just start painting versus a drywall guy who maybe doesn't, you know, you know.   Jason Hull (42:38) I like it.   John Casmon (42:55) mud the drywall properly or doesn't sand it down. So they got to do all this extra work before they start their process. Right. So a painter is going to know a great drywall guy. And in this case, it's really hard on ⁓ the property manager because you guys are the ones who do the work. But if you are looking for syndicators, OK, well syndicators, person who buys the deal. Well, who sells the deal? A broker. Find brokers. Go to a broker, commercial multifamily broker and ask them, hey,   Jason Hull (43:01) I love this.   Yeah.   John Casmon (43:25) Do you know some groups or you have properties that you're going to list? Here are the kind of deals we want to do now on the flip side of that. You got to be good at your job, right? You got to sell yourself and share what you do. So if you've got a great track record, a great resume, showcase that, bring that broker through and let them know, hey, we're looking to scale our property management business here. Here are the kind of assets that we want to manage. If you come across any of these that you're going to list, would you mind keeping our main name out there or referring us or giving us introductions to any of those buyers?   Jason Hull (43:53) Yeah.   John Casmon (43:54) so that we can throw our hat in the running to manage these properties. That's a phenomenal way to do that. And it allows you to shine and expand your relationships in your core networks and in your core markets.   Jason Hull (44:06) Brilliant. think I love the, I love Ashley's idea that you shared, you know, the drywall. Yeah. The painters, like they don't want to be painting over a crappy drywall. They're like, this is a mess. Like this doesn't even look good in my job. Now I'm going to look bad. Yeah. So the brokers know who maybe those best syndicators are. And so they could just go to the brokers and say, Hey, who's, who's doing deals like this? Who who's got things going on? Like who could you connect me with?   And I avoid maybe.   John Casmon (44:36) And on top of that, keep in mind, too, like what   are the times when? Yeah, but think about to like when is a property hiring or bringing on a new property manager? Right. So it's either a current owners firing the existing property manager or the property is being sold. Right. So, I mean, if you can get in during that transition phase, that's going to help you tremendously. And if even if they're firing their existing property manager, you can think through, OK, how do I?   Jason Hull (44:51) Yeah. Yeah.   John Casmon (45:06) work myself and get my name out there. And a lot of times, again, you're going to ask, right? You're going to ask other investors. If I were going through that process, I'm going to call my buddies into space, right? And say, hey, man, having a hard time, my current PM is not working out or we're not hitting our objectives, looking at some other options. Do you have any experience with these guys? What do you know about these guys? Or do you have anybody you could recommend? It's word of mouth, right? So that's what's going to start happening as well. So you kind of have to get out there and network and let folks know who you are, what you do. But you want to be someone who   people can say, yeah, these guys are amazing. You know, they, they only had an eight unit, but they crushed my eight unit for me. I'm sure they kill your 25 unit or your 50 unit. And you've got to start building that rapport and building your reputation in your market.   Jason Hull (45:44) Yeah.   Nice. This is good advice, my friend. So, cool. For those that maybe are investors listening to this show, ⁓ I'd love to hear a little bit about what you do, how you do run your syndication, and how they can ⁓ make things more passive, if that's what they're looking   John Casmon (46:08) Yeah, man. So there are lots of different ways to get in. If you are looking to be more passive, ⁓ high level, here's how it works. OK, so first and foremost, me and my team would go out. We look for the deals. We focus on a really tight radius. So we're in Cincinnati. We like Cincinnati, Columbus, Louisville, Kentucky. Really a two hour radius of the Cincinnati market is where we focus. And right now we actually think there's more opportunities locally. So we're really honed in on Cincinnati right now. But we focus on that once we find a deal.   We reach out to folks in our network. So we have folks in our investor list. ⁓ Once they're on our list, we kind of have a quick vetting process and then we can share opportunities with them. Once they see that opportunity, they get a chance to review it. We like to have a webinar where we answer any questions about the deal. I think for new investors, it's a great way to learn because we have a lot of experienced investors who ask very intelligent, thoughtful questions that   Many first time investors probably would not even think of. And that's a great way to learn, right? And ultimately when it comes to this space, it's really about education. know, it's educating yourself, understanding how you think about risk, how you mitigate risk in your investment choices. And those webinars are a great chance for you to learn about that the first time. Once you've done that, you can go ahead and fill out our official paperwork with our SEC documents.   Jason Hull (47:30) Mm-hmm.   John Casmon (47:30) And then   once you're through there, you can make the investment. But the first thing is just to get on our list, you can have access to the deals. And before you do that, we've actually put together a guide that can help people because I found that when I have these calls, people don't ask great questions. Sometimes they do. But I want to make sure that you are informed and well educated because this is a big investment. You know, this is not a 599 thing. And if it doesn't work out, OK, well, I just wasted six bucks. No.   Jason Hull (47:54) .   John Casmon (47:59) We're asking you to make a pretty large investment, whether it's with us or with others. If that's what you're looking to do, I want to make sure you're well informed. So we put together a guide. It's seven questions you must ask before investing in apartments. You can get that on our website. It's casmancapital.com slash seven questions, but it gets into questions around the market itself, the operating team, what you should be looking for, the deal. What is the story of this property? What's the business plan? And it helps you identify different levels of risk because the reality is   Anything can work, but you want to mitigate risk as much as possible, particularly when you're a passive investor, because you are basically saying, I'm trusting these people to find the right deal and execute. And you want to make sure that you are finding and identifying the right individuals who have a proven track record doing the thing that they are asking to do. When I hear about people losing money in real estate. At least 50, if not 70 % of the time.   Jason Hull (48:35) Hmm.   John Casmon (48:57) It is someone doing something for the first time. It is the first time in the market, first time doing this kind of deal, first time doing this kind of business plan. And. I can't tell you how frustrating it is because it's a big red flag, and it's not to say they can't do it and can't have success. But if it's your first time, I want to see how you're mitigating that right. You want to partner with someone who does have the experience you want. Like there are lot of things that you can do to put the odds in your favor. And when you're a passive investor.   Jason Hull (48:59) Mm, yeah.   John Casmon (49:26) It is not your job to hope. Your job is to analyze the information in front of you and make an informed decision. So this guide can help you do that.   Jason Hull (49:34) Yeah, love it. I'm going to run a quick word from our sponsor real quick. Our sponsor for this episode is Vendero. And many of you tell me that property management maintenance is probably the least enjoyable part of being a property manager and definitely the most time consuming. But what if you could cut that workload by up to 85 percent? That's exactly what Vendero has achieved. So they leverage cutting edge AI technology to handle nearly all your maintenance tasks from initiating work orders.   Troubleshooting, coordinating with vendors and reporting. This AI doesn't just automate, it becomes your ideal employee. Learning your preferences, executing tasks flawlessly and never needing a day off and never quitting. This frees you up to focus on the critical tasks that really move the needle for your business, whether that's refining operations, expanding your portfolio or even just taking a well-deserved break. Don't let maintenance drag you down. Step up your property management game with Vendero. Visit vendero.ai slash door grow today and make this the last maintenance hire you'll ever need.   All right, so John, this is super helpful. love you've got your list. ⁓ You got your webinar, you've got your guide. I would recommend property managers listening to this. If they're curious about the world of syndication, that they start getting into your stuff and seeing how an expert like you is doing this and maybe even get involved in some of the deals with you or something might be a good idea. And they can kind of get a feel for how this works. And then maybe they'll say, I don't want to do what John does.   And I'll just find people that do, but they'll at least understand how they could partner with people like that. then, or they may decide, you know what? John's clever, but I'm clever too. I might be able to figure out how to do this too. And maybe they'll do it too. And, but I think there's a solid opportunity for property managers that want to be in the multifamily space and do multifamily management to find third party people that are doing these syndication deals. They need good property managers and property managers want more doors and they want to grow.   And if you don't, because your business sucks and it's uncomfortable, then reach out to me. I'll help you out. We'll get you dialed in. But ⁓ John, what else would you say to the investors that are maybe they're familiar with this and they've done some real estate investing and they've worked with some syndications ⁓ and they get on your list to do the webinar. What would you say to them next?   John Casmon (51:56) Yeah, I think the biggest thing is understand what you're looking for. You know, I think one of the biggest challenges for investors is when you can't pull the trigger, it's typically because you haven't figured out what you're solving for. Are you looking for passive income? So you're just looking for a cash flow? Are you looking for long term wealth appreciation? Are you looking for tax benefits and to reduce kind of your tax liability? Do just want to diversify? Maybe you got feel like you have too much in a stock market, just like we put something somewhere else. So.   Figure out what you're actually solving for. Understand your risk tolerance, you know, because every deal is different. In our case, we do value add B class deals. That's a fancy way of just saying we like properties that already making money that are solid, solid tenant based. Think of when I say B class, I'm thinking of all stuff that was built maybe 30 years ago, maybe 40, maybe 20 years ago. Stuff that.   your teachers, your firefighters, your police officers, places where they might rent. So desirable locations, not luxury, not super high end, not, you know, super courts, everything. ⁓ But, you know, places that you would want your kid, your kid was in college, places you would be fine with your kid living, right? So you're thinking about that stuff. That's, you know, I don't say affordable stuff. That's not crazy price. So that's kind of what we focus on.   Jason Hull (53:15) So would   that be like, is that how you find the best markets then?   John Casmon (53:21) That's part of it. That's our strategy. There are different strategies that people utilize. I have found for us that is a sweet spot where we can take those kind of assets, modernize them and create value for potential renters. Some people like to focus only on they call it core plus right where they're buying newer stuff, stuff built five years ago or three years ago. And maybe it was, you know, leased up and they're just going to go in and hold it longer. You'll find other ways to add more money through amenities.   Jason Hull (53:35) Okay.   John Casmon (53:50) So some people do that strategy. Some people like older properties where they're buying more distressed or much older properties and are trying to fully renovate them and bring them up. There are strategies out there, something like new construction, stuff that doesn't exist. They want to build from the ground up. So it really comes down to you. Every investing strategy has a different level of risk. This has nothing to with real estate, right? This is investing in general. you're buying, you know, know, value stocks versus growth stocks versus Internet, it's the same stuff, right?   So you just have to figure out your level of risk. We like value at B-class multifamily deals. Once you understand your level of risk and balance that with your return expectations or projections, that's when you can figure out which investments actually make sense. You know, I have some folks who they like to invest in what we call trophy assets. And...   They may not know that right away, but when you send them a couple of deals and they look at the property like, ⁓ it's okay. They want something. They want something they can brag about. They want to drive you by like, see that building over there? That's me. And if that's fine, if that's what you want, understand what comes with that, right? That's going to be a lower term, right? Because these are, there's not much value to create, right? You've got a brand new property. It's A class, rents are $2,500. There's not a whole lot you can do there. And because of that,   Jason Hull (54:49) Yeah, they don't want to show that off. Look what I'm connecting.   OK, right.   Thank   Yeah.   John Casmon (55:13) There's not as much risk. So you're going to get less return because there's less risk. That's fun. Some people want to maximize their return, right? Hey, I don't need this money. I want to let it ride for 20 years. So they might want to do new construction or they might want to do a deep discount, highly distressed vacant property that needs, you know, $50,000 per unit to renovate it and turn around because the upside is there. So it just depends on that investor and your level of risk. Right. And most of us fall somewhere in the middle.   Jason Hull (55:27) Thank   John Casmon (55:43) which is kind of our strategy. figure out your level of risk tolerance, what you're looking for. And sometimes you don't know until you start looking at a Because you might think you're a cashflow person until I show you what cash flows. And you're like, oh, no, I don't want to be in that de

The Coalition unveils a plan to criminalise the helping of I-S group connected families, More nuclear talks between Iran and the U-S will go ahead amid threats of war, South Africa ends India's winning streak in the T20 World Cup.

Team Canada settles for silver after 2-1 overtime loss to U-S in Olympic men's hockey final. Olympic officials call for urgent funding boost as Canada's "pay-to-play" model hits a breaking point. Secret Service kills armed Mar-a-Lago intruder who was recently reported missing from North Carolina. Pakistan launches deadly retaliatory airstrikes in Afghanistan as Taliban warns of a "calculated response." Russia launches massive aerial barrage as Slovakia issues power ultimatum to Ukraine. TSA PreCheck is back online despite the ongoing shutdown.

Welcome to this fact-laden, episode of Light ‘Em Up.As we go to air - we're halfway through Black History Month.2026 marks the 100th anniversary of Black History Month, first established as "Negro History Week" by Dr. Carter G. Woodson in February 1926. The 2026 national theme is “A Century of Black History Commemorations”, honoring 100 years of recognizing the achievements and contributions of Black Americans to U.S. history.Black history is American history!We feel richly blessed and highly favored to sit down and visit, honoring Black History Month, with a dear friend of ours and a show favorite.  Dr. Sandy Womack, Jr.Dr. Womack Jr. has devoted his life to being a servant leader, educator and striving to be the best at whatever he attempts. He is an NCAA All-American wrestler with a doctorate in educational leadership, author, former principal, district administrator, equity trainer, motivational speaker, and much more.He is retired from his current role at the close of January (2026) after a generation (33+ years) of dedicating his life to urban education.This year's Black History Month arrives as our democracy is being tested in unprecedented ways, and the future of the republic hangs in the balance.Dr. Womack expressed grave concerns that we are “in an in between place similar to the times during the Reconstruction era after the Civil War (1865 – 1877) where the Federal Government sent in troops to ensure that the freed Blacks had and maintained the right to vote, to assemble, and the rights to a public education”.Based on reports, executive actions, and policy initiatives from 2025 and early 2026, the second Trump administration has pursued a broad agenda focused on rolling back diversity initiatives, changing civil rights enforcement, and altering educational and economic policies that critics argue disproportionately affect Black Americans.  And federal agents are arresting journalists (Don Lemon) and gunning down civilians in the streets.In this exclusive interview, we'll drill down on:— Assessing the pulse of the country — where are we as a nation and as people with the recent killings of Renee Good and Alex Pretti in Minneapolis?— The importance of taking an active role in the franchise (voting) how voting “changes policy, policy impacts practices and practices impact politics”.He feels deeply that the future “depends on the youth of today” — to see wide eyed the abuses and oversteps by this administration which will “serve as the catalyst to push the younger generation to the polls to vote in large and impressive ways”.He stressed the importance of “collectively coming together” and working to find some “communality” — at the end of the day, we have “more things that unite us than divide” — we sadly choose to focus and give attention to those aspects that are tribal and exacerbate division.We've strayed beyond an inflection point — democracy is actively in peril.“We have to become students of history. We have to read, write, speak and most importantly be able to think critically” to be educated and informed sufficiently to question the current status quo — for those who fail to learn from history are doomed to repeat it.He emphatically declares that: “Outcomes won't change until incomes change.” He urges all who will listen to “don't let your lying eyes fool you”.“The Party told you to reject the evidence of your eyes and ears. It was their final, most essential command.” ― George Orwell, 1984Tune in to our sponsors Newsly & Feedspot! We want to hear from you!Support the show

The Daily Quiz - Music Today's Questions: Question 1: Which American rapper, producer and actor released the studio album 'Encore'? Question 2: Which English rock band released the album 'Abbey Road'? Question 3: Which band did Zayn Malik belong to? Question 4: Which English rock band released the song 'Helter Skelter'? Question 5: Which singer released the song 'Run This Town'? Question 6: Which British musician collaborated with Queen on the song 'Under Pressure'? Question 7: Whose real name is Annie Mae Bullock? Question 8: Which English rock band released the song 'Us and Them'? This podcast is produced by Klassic Studios Learn more about your ad choices. Visit megaphone.fm/adchoices

Enjoy this special feed drop of our sister show 'In This Economy?!'Canadian restaurant operators could be in for another rough year – thanks to uncertainty around trade with the U-S and higher food costs – along with a few other factors.And the industry, which has been struggling for years, is important to the country's economy. But the numbers are not trending in the right direction.Hope is not lost.Host Kris McCusker takes a closer look at a report from Restaurants Canada with the President and CEO, Kelly Higginson. We love feedback at The Big Story, as well as suggestions for future episodes. You can find us:Through email at hello@thebigstorypodcast.ca Or @thebigstory.bsky.social on Bluesky

X: @GarrettInExile @americasrt1776 @ileaderssummit @NatashaSrdoc @JoelAnandUSA @supertalk Join America's Roundtable radio co-hosts Natasha Srdoc and Joel Anand Samy with the Honorable Thomas Garrett, Jr., member of the Commonwealth of Virginia's House of Delegates and former US Congressman. The conversation will focus on the state of America's economy, Trump's economic reforms, US-Iran Talks, America's ties with Israel, Virginia's radical changes with major tax hikes and sweeping gun control legislation and a new redistricting initiative which may leave Virginia's Congressional delegation with a 10-1 in favor of Democrats rather than the current 6-5 Republican edge. This could pose challenges in the mid-term elections.a leading attorney, currently serving as a legislator in the Commonwealth of Virginia, former Assistant Attorney General and former U.S. Congressman. The Washington Post's Editorial Board: "Brass-knuckled hypocrisy in Virginia" Quote: _The self-styled democracy party isn't behaving democratically. Democrats in Richmond are trying to effectively disenfranchise millions of Virginians by redrawing congressional maps to give themselves 10 of the commonwealth's 11 House seats — giving Democrats control of 91 percent of House seats in a state where Republicans lost the last presidential election by just six points. Most know better, including the governor. Abigail Spanberger was among the two-thirds of Virginians who voted in 2020 to transfer once-a-decade redistricting from the legislature to a bipartisan commission. “Gerrymandering is detrimental to our democracy,” she said back then. On Friday, Spanberger signed a bill to schedule an April 21 referendum that would move it back. The governor said it was necessary “to let voters respond to extreme measures taken by other states.”_ Bio | Tom Garrett Virginia Delegate Tom Garrett earned his Bachelor's degree from the University of Richmond. After the University of Richmond, Tom Garrett became an artillery officer in the United States Army. Achieving the rank of Captain, Tom led soldiers overseas—most notably while deployed in Bosnia. Upon returning to the States, Tom earned his J.D. from the University of Richmond and quickly became an Assistant Attorney General for Virginia. In 2016, Tom was elected to represent Virginia's 5th Congressional District in the United States House of Representatives. Tom won that election with the most votes ever in the 5th Congressional District. While in Congress, Tom served on the Foreign Affairs Committee, Homeland Security Committee, Education and Workforce Committee and was a member of the Freedom Caucus. An expert on Iran and the Middle East, Tom Garrett's analysis and insights are enlightening as America's foreign policy and national security concerns are focused on a strategic region adversely impacted by Iran, a state sponsor of terrorism. In the years since, Tom Garrett has dedicated his life to fighting for the oppressed and forgotten not only here in America, but around the world. Tom has been working on a global docu-series project, Exile, which tells the untold stories of those who are persecuted based on their faith or ethnicity. In addition to continuing his work as a defense attorney, Tom has served as a consultant and most recently, cofounder for firms working in global energy development. americasrt.com https://summitleadersusa.com/ | https://jerusalemleaderssummit.com/ America's Roundtable on Apple Podcasts: https://podcasts.apple.com/us/podcast/americas-roundtable/id1518878472 X: @GarrettInExile @americasrt1776 @ileaderssummit @NatashaSrdoc @JoelAnandUSA @supertalk America's Roundtable is co-hosted by Natasha Srdoc and Joel Anand Samy, co-founders of International Leaders Summit and the Jerusalem Leaders Summit. America's Roundtable radio program focuses on America's economy, healthcare reform, rule of law, security and trade, and its strategic partnership with rule of law nations around the world. The radio program features high-ranking US administration officials, cabinet members, members of Congress, state government officials, distinguished diplomats, business and media leaders and influential thinkers from around the world. Tune into America's Roundtable Radio program from Washington, DC via live streaming on Saturday mornings via 68 radio stations at 7:30 A.M. (ET) on Lanser Broadcasting Corporation covering the Michigan and the Midwest market, and at 7:30 A.M. (CT) on SuperTalk Mississippi — SuperTalk.FM reaching listeners in every county within the State of Mississippi, and neighboring states in the South including Alabama, Arkansas, Louisiana and Tennessee. Tune into WTON in Central Virginia on Sunday mornings at 6:00 A.M. (ET). Listen to America's Roundtable on digital platforms including Apple Podcasts, Spotify, Amazon, Google and other key online platforms. Listen live, Saturdays at 7:30 A.M. (CT) on SuperTalk | https://www.supertalk.fm

From rewriting Google's search stack in the early 2000s to reviving sparse trillion-parameter models and co-designing TPUs with frontier ML research, Jeff Dean has quietly shaped nearly every layer of the modern AI stack. As Chief AI Scientist at Google and a driving force behind Gemini, Jeff has lived through multiple scaling revolutions from CPUs and sharded indices to multimodal models that reason across text, video, and code.Jeff joins us to unpack what it really means to “own the Pareto frontier,” why distillation is the engine behind every Flash model breakthrough, how energy (in picojoules) not FLOPs is becoming the true bottleneck, what it was like leading the charge to unify all of Google's AI teams, and why the next leap won't come from bigger context windows alone, but from systems that give the illusion of attending to trillions of tokens.We discuss:* Jeff's early neural net thesis in 1990: parallel training before it was cool, why he believed scaling would win decades early, and the “bigger model, more data, better results” mantra that held for 15 years* The evolution of Google Search: sharding, moving the entire index into memory in 2001, softening query semantics pre-LLMs, and why retrieval pipelines already resemble modern LLM systems* Pareto frontier strategy: why you need both frontier “Pro” models and low-latency “Flash” models, and how distillation lets smaller models surpass prior generations* Distillation deep dive: ensembles → compression → logits as soft supervision, and why you need the biggest model to make the smallest one good* Latency as a first-class objective: why 10–50x lower latency changes UX entirely, and how future reasoning workloads will demand 10,000 tokens/sec* Energy-based thinking: picojoules per bit, why moving data costs 1000x more than a multiply, batching through the lens of energy, and speculative decoding as amortization* TPU co-design: predicting ML workloads 2–6 years out, speculative hardware features, precision reduction, sparsity, and the constant feedback loop between model architecture and silicon* Sparse models and “outrageously large” networks: trillions of parameters with 1–5% activation, and why sparsity was always the right abstraction* Unified vs. specialized models: abandoning symbolic systems, why general multimodal models tend to dominate vertical silos, and when vertical fine-tuning still makes sense* Long context and the illusion of scale: beyond needle-in-a-haystack benchmarks toward systems that narrow trillions of tokens to 117 relevant documents* Personalized AI: attending to your emails, photos, and documents (with permission), and why retrieval + reasoning will unlock deeply personal assistants* Coding agents: 50 AI interns, crisp specifications as a new core skill, and how ultra-low latency will reshape human–agent collaboration* Why ideas still matter: transformers, sparsity, RL, hardware, systems — scaling wasn't blind; the pieces had to multiply togetherShow Notes:* Gemma 3 Paper* Gemma 3* Gemini 2.5 Report* Jeff Dean's “Software Engineering Advice fromBuilding Large-Scale Distributed Systems” Presentation (with Back of the Envelope Calculations)* Latency Numbers Every Programmer Should Know by Jeff Dean* The Jeff Dean Facts* Jeff Dean Google Bio* Jeff Dean on “Important AI Trends” @Stanford AI Club* Jeff Dean & Noam Shazeer — 25 years at Google (Dwarkesh)—Jeff Dean* LinkedIn: https://www.linkedin.com/in/jeff-dean-8b212555* X: https://x.com/jeffdeanGoogle* https://google.com* https://deepmind.googleFull Video EpisodeTimestamps00:00:04 — Introduction: Alessio & Swyx welcome Jeff Dean, chief AI scientist at Google, to the Latent Space podcast00:00:30 — Owning the Pareto Frontier & balancing frontier vs low-latency models00:01:31 — Frontier models vs Flash models + role of distillation00:03:52 — History of distillation and its original motivation00:05:09 — Distillation's role in modern model scaling00:07:02 — Model hierarchy (Flash, Pro, Ultra) and distillation sources00:07:46 — Flash model economics & wide deployment00:08:10 — Latency importance for complex tasks00:09:19 — Saturation of some tasks and future frontier tasks00:11:26 — On benchmarks, public vs internal00:12:53 — Example long-context benchmarks & limitations00:15:01 — Long-context goals: attending to trillions of tokens00:16:26 — Realistic use cases beyond pure language00:18:04 — Multimodal reasoning and non-text modalities00:19:05 — Importance of vision & motion modalities00:20:11 — Video understanding example (extracting structured info)00:20:47 — Search ranking analogy for LLM retrieval00:23:08 — LLM representations vs keyword search00:24:06 — Early Google search evolution & in-memory index00:26:47 — Design principles for scalable systems00:28:55 — Real-time index updates & recrawl strategies00:30:06 — Classic “Latency numbers every programmer should know”00:32:09 — Cost of memory vs compute and energy emphasis00:34:33 — TPUs & hardware trade-offs for serving models00:35:57 — TPU design decisions & co-design with ML00:38:06 — Adapting model architecture to hardware00:39:50 — Alternatives: energy-based models, speculative decoding00:42:21 — Open research directions: complex workflows, RL00:44:56 — Non-verifiable RL domains & model evaluation00:46:13 — Transition away from symbolic systems toward unified LLMs00:47:59 — Unified models vs specialized ones00:50:38 — Knowledge vs reasoning & retrieval + reasoning00:52:24 — Vertical model specialization & modules00:55:21 — Token count considerations for vertical domains00:56:09 — Low resource languages & contextual learning00:59:22 — Origins: Dean's early neural network work01:10:07 — AI for coding & human–model interaction styles01:15:52 — Importance of crisp specification for coding agents01:19:23 — Prediction: personalized models & state retrieval01:22:36 — Token-per-second targets (10k+) and reasoning throughput01:23:20 — Episode conclusion and thanksTranscriptAlessio Fanelli [00:00:04]: Hey everyone, welcome to the Latent Space podcast. This is Alessio, founder of Kernel Labs, and I'm joined by Swyx, editor of Latent Space. Shawn Wang [00:00:11]: Hello, hello. We're here in the studio with Jeff Dean, chief AI scientist at Google. Welcome. Thanks for having me. It's a bit surreal to have you in the studio. I've watched so many of your talks, and obviously your career has been super legendary. So, I mean, congrats. I think the first thing must be said, congrats on owning the Pareto Frontier.Jeff Dean [00:00:30]: Thank you, thank you. Pareto Frontiers are good. It's good to be out there.Shawn Wang [00:00:34]: Yeah, I mean, I think it's a combination of both. You have to own the Pareto Frontier. You have to have like frontier capability, but also efficiency, and then offer that range of models that people like to use. And, you know, some part of this was started because of your hardware work. Some part of that is your model work, and I'm sure there's lots of secret sauce that you guys have worked on cumulatively. But, like, it's really impressive to see it all come together in, like, this slittily advanced.Jeff Dean [00:01:04]: Yeah, yeah. I mean, I think, as you say, it's not just one thing. It's like a whole bunch of things up and down the stack. And, you know, all of those really combine to help make UNOS able to make highly capable large models, as well as, you know, software techniques to get those large model capabilities into much smaller, lighter weight models that are, you know, much more cost effective and lower latency, but still, you know, quite capable for their size. Yeah.Alessio Fanelli [00:01:31]: How much pressure do you have on, like, having the lower bound of the Pareto Frontier, too? I think, like, the new labs are always trying to push the top performance frontier because they need to raise more money and all of that. And you guys have billions of users. And I think initially when you worked on the CPU, you were thinking about, you know, if everybody that used Google, we use the voice model for, like, three minutes a day, they were like, you need to double your CPU number. Like, what's that discussion today at Google? Like, how do you prioritize frontier versus, like, we have to do this? How do we actually need to deploy it if we build it?Jeff Dean [00:02:03]: Yeah, I mean, I think we always want to have models that are at the frontier or pushing the frontier because I think that's where you see what capabilities now exist that didn't exist at the sort of slightly less capable last year's version or last six months ago version. At the same time, you know, we know those are going to be really useful for a bunch of use cases, but they're going to be a bit slower and a bit more expensive than people might like for a bunch of other broader models. So I think what we want to do is always have kind of a highly capable sort of affordable model that enables a whole bunch of, you know, lower latency use cases. People can use them for agentic coding much more readily and then have the high-end, you know, frontier model that is really useful for, you know, deep reasoning, you know, solving really complicated math problems, those kinds of things. And it's not that. One or the other is useful. They're both useful. So I think we'd like to do both. And also, you know, through distillation, which is a key technique for making the smaller models more capable, you know, you have to have the frontier model in order to then distill it into your smaller model. So it's not like an either or choice. You sort of need that in order to actually get a highly capable, more modest size model. Yeah.Alessio Fanelli [00:03:24]: I mean, you and Jeffrey came up with the solution in 2014.Jeff Dean [00:03:28]: Don't forget, L'Oreal Vinyls as well. Yeah, yeah.Alessio Fanelli [00:03:30]: A long time ago. But like, I'm curious how you think about the cycle of these ideas, even like, you know, sparse models and, you know, how do you reevaluate them? How do you think about in the next generation of model, what is worth revisiting? Like, yeah, they're just kind of like, you know, you worked on so many ideas that end up being influential, but like in the moment, they might not feel that way necessarily. Yeah.Jeff Dean [00:03:52]: I mean, I think distillation was originally motivated because we were seeing that we had a very large image data set at the time, you know, 300 million images that we could train on. And we were seeing that if you create specialists for different subsets of those image categories, you know, this one's going to be really good at sort of mammals, and this one's going to be really good at sort of indoor room scenes or whatever, and you can cluster those categories and train on an enriched stream of data after you do pre-training on a much broader set of images. You get much better performance. If you then treat that whole set of maybe 50 models you've trained as a large ensemble, but that's not a very practical thing to serve, right? So distillation really came about from the idea of, okay, what if we want to actually serve that and train all these independent sort of expert models and then squish it into something that actually fits in a form factor that you can actually serve? And that's, you know, not that different from what we're doing today. You know, often today we're instead of having an ensemble of 50 models. We're having a much larger scale model that we then distill into a much smaller scale model.Shawn Wang [00:05:09]: Yeah. A part of me also wonders if distillation also has a story with the RL revolution. So let me maybe try to articulate what I mean by that, which is you can, RL basically spikes models in a certain part of the distribution. And then you have to sort of, well, you can spike models, but usually sometimes... It might be lossy in other areas and it's kind of like an uneven technique, but you can probably distill it back and you can, I think that the sort of general dream is to be able to advance capabilities without regressing on anything else. And I think like that, that whole capability merging without loss, I feel like it's like, you know, some part of that should be a distillation process, but I can't quite articulate it. I haven't seen much papers about it.Jeff Dean [00:06:01]: Yeah, I mean, I tend to think of one of the key advantages of distillation is that you can have a much smaller model and you can have a very large, you know, training data set and you can get utility out of making many passes over that data set because you're now getting the logits from the much larger model in order to sort of coax the right behavior out of the smaller model that you wouldn't otherwise get with just the hard labels. And so, you know, I think that's what we've observed. Is you can get, you know, very close to your largest model performance with distillation approaches. And that seems to be, you know, a nice sweet spot for a lot of people because it enables us to kind of, for multiple Gemini generations now, we've been able to make the sort of flash version of the next generation as good or even substantially better than the previous generations pro. And I think we're going to keep trying to do that because that seems like a good trend to follow.Shawn Wang [00:07:02]: So, Dara asked, so it was the original map was Flash Pro and Ultra. Are you just sitting on Ultra and distilling from that? Is that like the mother load?Jeff Dean [00:07:12]: I mean, we have a lot of different kinds of models. Some are internal ones that are not necessarily meant to be released or served. Some are, you know, our pro scale model and we can distill from that as well into our Flash scale model. So I think, you know, it's an important set of capabilities to have and also inference time scaling. It can also be a useful thing to improve the capabilities of the model.Shawn Wang [00:07:35]: And yeah, yeah, cool. Yeah. And obviously, I think the economy of Flash is what led to the total dominance. I think the latest number is like 50 trillion tokens. I don't know. I mean, obviously, it's changing every day.Jeff Dean [00:07:46]: Yeah, yeah. But, you know, by market share, hopefully up.Shawn Wang [00:07:50]: No, I mean, there's no I mean, there's just the economics wise, like because Flash is so economical, like you can use it for everything. Like it's in Gmail now. It's in YouTube. Like it's yeah. It's in everything.Jeff Dean [00:08:02]: We're using it more in our search products of various AI mode reviews.Shawn Wang [00:08:05]: Oh, my God. Flash past the AI mode. Oh, my God. Yeah, that's yeah, I didn't even think about that.Jeff Dean [00:08:10]: I mean, I think one of the things that is quite nice about the Flash model is not only is it more affordable, it's also a lower latency. And I think latency is actually a pretty important characteristic for these models because we're going to want models to do much more complicated things that are going to involve, you know, generating many more tokens from when you ask the model to do so. So, you know, if you're going to ask the model to do something until it actually finishes what you ask it to do, because you're going to ask now, not just write me a for loop, but like write me a whole software package to do X or Y or Z. And so having low latency systems that can do that seems really important. And Flash is one direction, one way of doing that. You know, obviously our hardware platforms enable a bunch of interesting aspects of our, you know, serving stack as well, like TPUs, the interconnect between. Chips on the TPUs is actually quite, quite high performance and quite amenable to, for example, long context kind of attention operations, you know, having sparse models with lots of experts. These kinds of things really, really matter a lot in terms of how do you make them servable at scale.Alessio Fanelli [00:09:19]: Yeah. Does it feel like there's some breaking point for like the proto Flash distillation, kind of like one generation delayed? I almost think about almost like the capability as a. In certain tasks, like the pro model today is a saturated, some sort of task. So next generation, that same task will be saturated at the Flash price point. And I think for most of the things that people use models for at some point, the Flash model in two generation will be able to do basically everything. And how do you make it economical to like keep pushing the pro frontier when a lot of the population will be okay with the Flash model? I'm curious how you think about that.Jeff Dean [00:09:59]: I mean, I think that's true. If your distribution of what people are asking people, the models to do is stationary, right? But I think what often happens is as the models become more capable, people ask them to do more, right? So, I mean, I think this happens in my own usage. Like I used to try our models a year ago for some sort of coding task, and it was okay at some simpler things, but wouldn't do work very well for more complicated things. And since then, we've improved dramatically on the more complicated coding tasks. And now I'll ask it to do much more complicated things. And I think that's true, not just of coding, but of, you know, now, you know, can you analyze all the, you know, renewable energy deployments in the world and give me a report on solar panel deployment or whatever. That's a very complicated, you know, more complicated task than people would have asked a year ago. And so you are going to want more capable models to push the frontier in the absence of what people ask the models to do. And that also then gives us. Insight into, okay, where does the, where do things break down? How can we improve the model in these, these particular areas, uh, in order to sort of, um, make the next generation even better.Alessio Fanelli [00:11:11]: Yeah. Are there any benchmarks or like test sets they use internally? Because it's almost like the same benchmarks get reported every time. And it's like, all right, it's like 99 instead of 97. Like, how do you have to keep pushing the team internally to it? Or like, this is what we're building towards. Yeah.Jeff Dean [00:11:26]: I mean, I think. Benchmarks, particularly external ones that are publicly available. Have their utility, but they often kind of have a lifespan of utility where they're introduced and maybe they're quite hard for current models. You know, I, I like to think of the best kinds of benchmarks are ones where the initial scores are like 10 to 20 or 30%, maybe, but not higher. And then you can sort of work on improving that capability for, uh, whatever it is, the benchmark is trying to assess and get it up to like 80, 90%, whatever. I, I think once it hits kind of 95% or something, you get very diminishing returns from really focusing on that benchmark, cuz it's sort of, it's either the case that you've now achieved that capability, or there's also the issue of leakage in public data or very related kind of data being, being in your training data. Um, so we have a bunch of held out internal benchmarks that we really look at where we know that wasn't represented in the training data at all. There are capabilities that we want the model to have. Um, yeah. Yeah. Um, that it doesn't have now, and then we can work on, you know, assessing, you know, how do we make the model better at these kinds of things? Is it, we need different kind of data to train on that's more specialized for this particular kind of task. Do we need, um, you know, a bunch of, uh, you know, architectural improvements or some sort of, uh, model capability improvements, you know, what would help make that better?Shawn Wang [00:12:53]: Is there, is there such an example that you, uh, a benchmark inspired in architectural improvement? Like, uh, I'm just kind of. Jumping on that because you just.Jeff Dean [00:13:02]: Uh, I mean, I think some of the long context capability of the, of the Gemini models that came, I guess, first in 1.5 really were about looking at, okay, we want to have, um, you know,Shawn Wang [00:13:15]: immediately everyone jumped to like completely green charts of like, everyone had, I was like, how did everyone crack this at the same time? Right. Yeah. Yeah.Jeff Dean [00:13:23]: I mean, I think, um, and once you're set, I mean, as you say that needed single needle and a half. Hey, stack benchmark is really saturated for at least context links up to 1, 2 and K or something. Don't actually have, you know, much larger than 1, 2 and 8 K these days or two or something. We're trying to push the frontier of 1 million or 2 million context, which is good because I think there are a lot of use cases where. Yeah. You know, putting a thousand pages of text or putting, you know, multiple hour long videos and the context and then actually being able to make use of that as useful. Try to, to explore the über graduation are fairly large. But the single needle in a haystack benchmark is sort of saturated. So you really want more complicated, sort of multi-needle or more realistic, take all this content and produce this kind of answer from a long context that sort of better assesses what it is people really want to do with long context. Which is not just, you know, can you tell me the product number for this particular thing?Shawn Wang [00:14:31]: Yeah, it's retrieval. It's retrieval within machine learning. It's interesting because I think the more meta level I'm trying to operate at here is you have a benchmark. You're like, okay, I see the architectural thing I need to do in order to go fix that. But should you do it? Because sometimes that's an inductive bias, basically. It's what Jason Wei, who used to work at Google, would say. Exactly the kind of thing. Yeah, you're going to win. Short term. Longer term, I don't know if that's going to scale. You might have to undo that.Jeff Dean [00:15:01]: I mean, I like to sort of not focus on exactly what solution we're going to derive, but what capability would you want? And I think we're very convinced that, you know, long context is useful, but it's way too short today. Right? Like, I think what you would really want is, can I attend to the internet while I answer my question? Right? But that's not going to happen. I think that's going to be solved by purely scaling the existing solutions, which are quadratic. So a million tokens kind of pushes what you can do. You're not going to do that to a trillion tokens, let alone, you know, a billion tokens, let alone a trillion. But I think if you could give the illusion that you can attend to trillions of tokens, that would be amazing. You'd find all kinds of uses for that. You would have attend to the internet. You could attend to the pixels of YouTube and the sort of deeper representations that we can find. You could attend to the form for a single video, but across many videos, you know, on a personal Gemini level, you could attend to all of your personal state with your permission. So like your emails, your photos, your docs, your plane tickets you have. I think that would be really, really useful. And the question is, how do you get algorithmic improvements and system level improvements that get you to something where you actually can attend to trillions of tokens? Right. In a meaningful way. Yeah.Shawn Wang [00:16:26]: But by the way, I think I did some math and it's like, if you spoke all day, every day for eight hours a day, you only generate a maximum of like a hundred K tokens, which like very comfortably fits.Jeff Dean [00:16:38]: Right. But if you then say, okay, I want to be able to understand everything people are putting on videos.Shawn Wang [00:16:46]: Well, also, I think that the classic example is you start going beyond language into like proteins and whatever else is extremely information dense. Yeah. Yeah.Jeff Dean [00:16:55]: I mean, I think one of the things about Gemini's multimodal aspects is we've always wanted it to be multimodal from the start. And so, you know, that sometimes to people means text and images and video sort of human-like and audio, audio, human-like modalities. But I think it's also really useful to have Gemini know about non-human modalities. Yeah. Like LIDAR sensor data from. Yes. Say, Waymo vehicles or. Like robots or, you know, various kinds of health modalities, x-rays and MRIs and imaging and genomics information. And I think there's probably hundreds of modalities of data where you'd like the model to be able to at least be exposed to the fact that this is an interesting modality and has certain meaning in the world. Where even if you haven't trained on all the LIDAR data or MRI data, you could have, because maybe that's not, you know, it doesn't make sense in terms of trade-offs of. You know, what you include in your main pre-training data mix, at least including a little bit of it is actually quite useful. Yeah. Because it sort of tempts the model that this is a thing.Shawn Wang [00:18:04]: Yeah. Do you believe, I mean, since we're on this topic and something I just get to ask you all the questions I always wanted to ask, which is fantastic. Like, are there some king modalities, like modalities that supersede all the other modalities? So a simple example was Vision can, on a pixel level, encode text. And DeepSeq had this DeepSeq CR paper that did that. Vision. And Vision has also been shown to maybe incorporate audio because you can do audio spectrograms and that's, that's also like a Vision capable thing. Like, so, so maybe Vision is just the king modality and like. Yeah.Jeff Dean [00:18:36]: I mean, Vision and Motion are quite important things, right? Motion. Well, like video as opposed to static images, because I mean, there's a reason evolution has evolved eyes like 23 independent ways, because it's such a useful capability for sensing the world around you, which is really what we want these models to be. So I think the only thing that we can be able to do is interpret the things we're seeing or the things we're paying attention to and then help us in using that information to do things. Yeah.Shawn Wang [00:19:05]: I think motion, you know, I still want to shout out, I think Gemini, still the only native video understanding model that's out there. So I use it for YouTube all the time. Nice.Jeff Dean [00:19:15]: Yeah. Yeah. I mean, it's actually, I think people kind of are not necessarily aware of what the Gemini models can actually do. Yeah. Like I have an example I've used in one of my talks. It had like, it was like a YouTube highlight video of 18 memorable sports moments across the last 20 years or something. So it has like Michael Jordan hitting some jump shot at the end of the finals and, you know, some soccer goals and things like that. And you can literally just give it the video and say, can you please make me a table of what all these different events are? What when the date is when they happened? And a short description. And so you get like now an 18 row table of that information extracted from the video, which is, you know, not something most people think of as like a turn video into sequel like table.Alessio Fanelli [00:20:11]: Has there been any discussion inside of Google of like, you mentioned tending to the whole internet, right? Google, it's almost built because a human cannot tend to the whole internet and you need some sort of ranking to find what you need. Yep. That ranking is like much different for an LLM because you can expect a person to look at maybe the first five, six links in a Google search versus for an LLM. Should you expect to have 20 links that are highly relevant? Like how do you internally figure out, you know, how do we build the AI mode that is like maybe like much broader search and span versus like the more human one? Yeah.Jeff Dean [00:20:47]: I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. With a giant number of web pages in our index, many of them are not relevant. So you identify a subset of them that are relevant with very lightweight kinds of methods. You know, you're down to like 30,000 documents or something. And then you gradually refine that to apply more and more sophisticated algorithms and more and more sophisticated sort of signals of various kinds in order to get down to ultimately what you show, which is, you know, the final 10 results or, you know, 10 results plus. Other kinds of information. And I think an LLM based system is not going to be that dissimilar, right? You're going to attend to trillions of tokens, but you're going to want to identify, you know, what are the 30,000 ish documents that are with the, you know, maybe 30 million interesting tokens. And then how do you go from that into what are the 117 documents I really should be paying attention to in order to carry out the tasks that the user has asked? And I think, you know, you can imagine systems where you have, you know, a lot of highly parallel processing to identify those initial 30,000 candidates, maybe with very lightweight kinds of models. Then you have some system that sort of helps you narrow down from 30,000 to the 117 with maybe a little bit more sophisticated model or set of models. And then maybe the final model is the thing that looks. So the 117 things that might be your most capable model. So I think it has to, it's going to be some system like that, that is really enables you to give the illusion of attending to trillions of tokens. Sort of the way Google search gives you, you know, not the illusion, but you are searching the internet, but you're finding, you know, a very small subset of things that are, that are relevant.Shawn Wang [00:22:47]: Yeah. I often tell a lot of people that are not steeped in like Google search history that, well, you know, like Bert was. Like he was like basically immediately inside of Google search and that improves results a lot, right? Like I don't, I don't have any numbers off the top of my head, but like, I'm sure you guys, that's obviously the most important numbers to Google. Yeah.Jeff Dean [00:23:08]: I mean, I think going to an LLM based representation of text and words and so on enables you to get out of the explicit hard notion of, of particular words having to be on the page, but really getting at the notion of this topic of this page or this page. Paragraph is highly relevant to this query. Yeah.Shawn Wang [00:23:28]: I don't think people understand how much LLMs have taken over all these very high traffic system, very high traffic. Yeah. Like it's Google, it's YouTube. YouTube has this like semantics ID thing where it's just like every token or every item in the vocab is a YouTube video or something that predicts the video using a code book, which is absurd to me for YouTube size.Jeff Dean [00:23:50]: And then most recently GROK also for, for XAI, which is like, yeah. I mean, I'll call out even before LLMs were used extensively in search, we put a lot of emphasis on softening the notion of what the user actually entered into the query.Shawn Wang [00:24:06]: So do you have like a history of like, what's the progression? Oh yeah.Jeff Dean [00:24:09]: I mean, I actually gave a talk in, uh, I guess, uh, web search and data mining conference in 2009, uh, where we never actually published any papers about the origins of Google search, uh, sort of, but we went through sort of four or five or six. generations, four or five or six generations of, uh, redesigning of the search and retrieval system, uh, from about 1999 through 2004 or five. And that talk is really about that evolution. And one of the things that really happened in 2001 was we were sort of working to scale the system in multiple dimensions. So one is we wanted to make our index bigger, so we could retrieve from a larger index, which always helps your quality in general. Uh, because if you don't have the page in your index, you're going to not do well. Um, and then we also needed to scale our capacity because we were, our traffic was growing quite extensively. Um, and so we had, you know, a sharded system where you have more and more shards as the index grows, you have like 30 shards. And then if you want to double the index size, you make 60 shards so that you can bound the latency by which you respond for any particular user query. Um, and then as traffic grows, you add, you add more and more replicas of each of those. And so we eventually did the math that realized that in a data center where we had say 60 shards and, um, you know, 20 copies of each shard, we now had 1200 machines, uh, with disks. And we did the math and we're like, Hey, one copy of that index would actually fit in memory across 1200 machines. So in 2001, we introduced, uh, we put our entire index in memory and what that enabled from a quality perspective was amazing. Um, and so we had more and more replicas of each of those. Before you had to be really careful about, you know, how many different terms you looked at for a query, because every one of them would involve a disk seek on every one of the 60 shards. And so you, as you make your index bigger, that becomes even more inefficient. But once you have the whole index in memory, it's totally fine to have 50 terms you throw into the query from the user's original three or four word query, because now you can add synonyms like restaurant and restaurants and cafe and, uh, you know, things like that. Uh, bistro and all these things. And you can suddenly start, uh, sort of really, uh, getting at the meaning of the word as opposed to the exact semantic form the user typed in. And that was, you know, 2001, very much pre LLM, but really it was about softening the, the strict definition of what the user typed in order to get at the meaning.Alessio Fanelli [00:26:47]: What are like principles that you use to like design the systems, especially when you have, I mean, in 2001, the internet is like. Doubling, tripling every year in size is not like, uh, you know, and I think today you kind of see that with LLMs too, where like every year the jumps in size and like capabilities are just so big. Are there just any, you know, principles that you use to like, think about this? Yeah.Jeff Dean [00:27:08]: I mean, I think, uh, you know, first, whenever you're designing a system, you want to understand what are the sort of design parameters that are going to be most important in designing that, you know? So, you know, how many queries per second do you need to handle? How big is the internet? How big is the index you need to handle? How much data do you need to keep for every document in the index? How are you going to look at it when you retrieve things? Um, what happens if traffic were to double or triple, you know, will that system work well? And I think a good design principle is you're going to want to design a system so that the most important characteristics could scale by like factors of five or 10, but probably not beyond that because often what happens is if you design a system for X. And something suddenly becomes a hundred X, that would enable a very different point in the design space that would not make sense at X. But all of a sudden at a hundred X makes total sense. So like going from a disk space index to a in memory index makes a lot of sense once you have enough traffic, because now you have enough replicas of the sort of state on disk that those machines now actually can hold, uh, you know, a full copy of the, uh, index and memory. Yeah. And that all of a sudden enabled. A completely different design that wouldn't have been practical before. Yeah. Um, so I'm, I'm a big fan of thinking through designs in your head, just kind of playing with the design space a little before you actually do a lot of writing of code. But, you know, as you said, in the early days of Google, we were growing the index, uh, quite extensively. We were growing the update rate of the index. So the update rate actually is the parameter that changed the most. Surprising. So it used to be once a month.Shawn Wang [00:28:55]: Yeah.Jeff Dean [00:28:56]: And then we went to a system that could update any particular page in like sub one minute. Okay.Shawn Wang [00:29:02]: Yeah. Because this is a competitive advantage, right?Jeff Dean [00:29:04]: Because all of a sudden news related queries, you know, if you're, if you've got last month's news index, it's not actually that useful for.Shawn Wang [00:29:11]: News is a special beast. Was there any, like you could have split it onto a separate system.Jeff Dean [00:29:15]: Well, we did. We launched a Google news product, but you also want news related queries that people type into the main index to also be sort of updated.Shawn Wang [00:29:23]: So, yeah, it's interesting. And then you have to like classify whether the page is, you have to decide which pages should be updated and what frequency. Oh yeah.Jeff Dean [00:29:30]: There's a whole like, uh, system behind the scenes that's trying to decide update rates and importance of the pages. So even if the update rate seems low, you might still want to recrawl important pages quite often because, uh, the likelihood they change might be low, but the value of having updated is high.Shawn Wang [00:29:50]: Yeah, yeah, yeah, yeah. Uh, well, you know, yeah. This, uh, you know, mention of latency and, and saving things to this reminds me of one of your classics, which I have to bring up, which is latency numbers. Every programmer should know, uh, was there a, was it just a, just a general story behind that? Did you like just write it down?Jeff Dean [00:30:06]: I mean, this has like sort of eight or 10 different kinds of metrics that are like, how long does a cache mistake? How long does branch mispredict take? How long does a reference domain memory take? How long does it take to send, you know, a packet from the U S to the Netherlands or something? Um,Shawn Wang [00:30:21]: why Netherlands, by the way, or is it, is that because of Chrome?Jeff Dean [00:30:25]: Uh, we had a data center in the Netherlands, um, so, I mean, I think this gets to the point of being able to do the back of the envelope calculations. So these are sort of the raw ingredients of those, and you can use them to say, okay, well, if I need to design a system to do image search and thumb nailing or something of the result page, you know, how, what I do that I could pre-compute the image thumbnails. I could like. Try to thumbnail them on the fly from the larger images. What would that do? How much dis bandwidth than I need? How many des seeks would I do? Um, and you can sort of actually do thought experiments in, you know, 30 seconds or a minute with the sort of, uh, basic, uh, basic numbers at your fingertips. Uh, and then as you sort of build software using higher level libraries, you kind of want to develop the same intuitions for how long does it take to, you know, look up something in this particular kind of.Shawn Wang [00:31:21]: I'll see you next time.Shawn Wang [00:31:51]: Which is a simple byte conversion. That's nothing interesting. I wonder if you have any, if you were to update your...Jeff Dean [00:31:58]: I mean, I think it's really good to think about calculations you're doing in a model, either for training or inference.Jeff Dean [00:32:09]: Often a good way to view that is how much state will you need to bring in from memory, either like on-chip SRAM or HBM from the accelerator. Attached memory or DRAM or over the network. And then how expensive is that data motion relative to the cost of, say, an actual multiply in the matrix multiply unit? And that cost is actually really, really low, right? Because it's order, depending on your precision, I think it's like sub one picodule.Shawn Wang [00:32:50]: Oh, okay. You measure it by energy. Yeah. Yeah.Jeff Dean [00:32:52]: Yeah. I mean, it's all going to be about energy and how do you make the most energy efficient system. And then moving data from the SRAM on the other side of the chip, not even off the off chip, but on the other side of the same chip can be, you know, a thousand picodules. Oh, yeah. And so all of a sudden, this is why your accelerators require batching. Because if you move, like, say, the parameter of a model from SRAM on the, on the chip into the multiplier unit, that's going to cost you a thousand picodules. So you better make use of that, that thing that you moved many, many times with. So that's where the batch dimension comes in. Because all of a sudden, you know, if you have a batch of 256 or something, that's not so bad. But if you have a batch of one, that's really not good.Shawn Wang [00:33:40]: Yeah. Yeah. Right.Jeff Dean [00:33:41]: Because then you paid a thousand picodules in order to do your one picodule multiply.Shawn Wang [00:33:46]: I have never heard an energy-based analysis of batching.Jeff Dean [00:33:50]: Yeah. I mean, that's why people batch. Yeah. Ideally, you'd like to use batch size one because the latency would be great.Shawn Wang [00:33:56]: The best latency.Jeff Dean [00:33:56]: But the energy cost and the compute cost inefficiency that you get is quite large. So, yeah.Shawn Wang [00:34:04]: Is there a similar trick like, like, like you did with, you know, putting everything in memory? Like, you know, I think obviously NVIDIA has caused a lot of waves with betting very hard on SRAM with Grok. I wonder if, like, that's something that you already saw with, with the TPUs, right? Like that, that you had to. Uh, to serve at your scale, uh, you probably sort of saw that coming. Like what, what, what hardware, uh, innovations or insights were formed because of what you're seeing there?Jeff Dean [00:34:33]: Yeah. I mean, I think, you know, TPUs have this nice, uh, sort of regular structure of 2D or 3D meshes with a bunch of chips connected. Yeah. And each one of those has HBM attached. Um, I think for serving some kinds of models, uh, you know, you, you pay a lot higher cost. Uh, and time latency, um, bringing things in from HBM than you do bringing them in from, uh, SRAM on the chip. So if you have a small enough model, you can actually do model parallelism, spread it out over lots of chips and you actually get quite good throughput improvements and latency improvements from doing that. And so you're now sort of striping your smallish scale model over say 16 or 64 chips. Uh, but as if you do that and it all fits in. In SRAM, uh, that can be a big win. So yeah, that's not a surprise, but it is a good technique.Alessio Fanelli [00:35:27]: Yeah. What about the TPU design? Like how much do you decide where the improvements have to go? So like, this is like a good example of like, is there a way to bring the thousand picojoules down to 50? Like, is it worth designing a new chip to do that? The extreme is like when people say, oh, you should burn the model on the ASIC and that's kind of like the most extreme thing. How much of it? Is it worth doing an hardware when things change so quickly? Like what was the internal discussion? Yeah.Jeff Dean [00:35:57]: I mean, we, we have a lot of interaction between say the TPU chip design architecture team and the sort of higher level modeling, uh, experts, because you really want to take advantage of being able to co-design what should future TPUs look like based on where we think the sort of ML research puck is going, uh, in some sense, because, uh, you know, as a hardware designer for ML and in particular, you're trying to design a chip starting today and that design might take two years before it even lands in a data center. And then it has to sort of be a reasonable lifetime of the chip to take you three, four or five years. So you're trying to predict two to six years out where, what ML computations will people want to run two to six years out in a very fast changing field. And so having people with interest. Interesting ML research ideas of things we think will start to work in that timeframe or will be more important in that timeframe, uh, really enables us to then get, you know, interesting hardware features put into, you know, TPU N plus two, where TPU N is what we have today.Shawn Wang [00:37:10]: Oh, the cycle time is plus two.Jeff Dean [00:37:12]: Roughly. Wow. Because, uh, I mean, sometimes you can squeeze some changes into N plus one, but, you know, bigger changes are going to require the chip. Yeah. Design be earlier in its lifetime design process. Um, so whenever we can do that, it's generally good. And sometimes you can put in speculative features that maybe won't cost you much chip area, but if it works out, it would make something, you know, 10 times as fast. And if it doesn't work out, well, you burned a little bit of tiny amount of your chip area on that thing, but it's not that big a deal. Uh, sometimes it's a very big change and we want to be pretty sure this is going to work out. So we'll do like lots of carefulness. Uh, ML experimentation to show us, uh, this is actually the, the way we want to go. Yeah.Alessio Fanelli [00:37:58]: Is there a reverse of like, we already committed to this chip design so we can not take the model architecture that way because it doesn't quite fit?Jeff Dean [00:38:06]: Yeah. I mean, you, you definitely have things where you're going to adapt what the model architecture looks like so that they're efficient on the chips that you're going to have for both training and inference of that, of that, uh, generation of model. So I think it kind of goes both ways. Um, you know, sometimes you can take advantage of, you know, lower precision things that are coming in a future generation. So you can, might train it at that lower precision, even if the current generation doesn't quite do that. Mm.Shawn Wang [00:38:40]: Yeah. How low can we go in precision?Jeff Dean [00:38:43]: Because people are saying like ternary is like, uh, yeah, I mean, I'm a big fan of very low precision because I think that gets, that saves you a tremendous amount of time. Right. Because it's picojoules per bit that you're transferring and reducing the number of bits is a really good way to, to reduce that. Um, you know, I think people have gotten a lot of luck, uh, mileage out of having very low bit precision things, but then having scaling factors that apply to a whole bunch of, uh, those, those weights. Scaling. How does it, how does it, okay.Shawn Wang [00:39:15]: Interesting. You, so low, low precision, but scaled up weights. Yeah. Huh. Yeah. Never considered that. Yeah. Interesting. Uh, w w while we're on this topic, you know, I think there's a lot of, um, uh, this, the concept of precision at all is weird when we're sampling, you know, uh, we just, at the end of this, we're going to have all these like chips that I'll do like very good math. And then we're just going to throw a random number generator at the start. So, I mean, there's a movement towards, uh, energy based, uh, models and processors. I'm just curious if you've, obviously you've thought about it, but like, what's your commentary?Jeff Dean [00:39:50]: Yeah. I mean, I think. There's a bunch of interesting trends though. Energy based models is one, you know, diffusion based models, which don't sort of sequentially decode tokens is another, um, you know, speculative decoding is a way that you can get sort of an equivalent, very small.Shawn Wang [00:40:06]: Draft.Jeff Dean [00:40:07]: Batch factor, uh, for like you predict eight tokens out and that enables you to sort of increase the effective batch size of what you're doing by a factor of eight, even, and then you maybe accept five or six of those tokens. So you get. A five, a five X improvement in the amortization of moving weights, uh, into the multipliers to do the prediction for the, the tokens. So these are all really good techniques and I think it's really good to look at them from the lens of, uh, energy, real energy, not energy based models, um, and, and also latency and throughput, right? If you look at things from that lens, that sort of guides you to. Two solutions that are gonna be, uh, you know, better from, uh, you know, being able to serve larger models or, you know, equivalent size models more cheaply and with lower latency.Shawn Wang [00:41:03]: Yeah. Well, I think, I think I, um, it's appealing intellectually, uh, haven't seen it like really hit the mainstream, but, um, I do think that, uh, there's some poetry in the sense that, uh, you know, we don't have to do, uh, a lot of shenanigans if like we fundamentally. Design it into the hardware. Yeah, yeah.Jeff Dean [00:41:23]: I mean, I think there's still a, there's also sort of the more exotic things like analog based, uh, uh, computing substrates as opposed to digital ones. Uh, I'm, you know, I think those are super interesting cause they can be potentially low power. Uh, but I think you often end up wanting to interface that with digital systems and you end up losing a lot of the power advantages in the digital to analog and analog to digital conversions. You end up doing, uh, at the sort of boundaries. And periphery of that system. Um, I still think there's a tremendous distance we can go from where we are today in terms of energy efficiency with sort of, uh, much better and specialized hardware for the models we care about.Shawn Wang [00:42:05]: Yeah.Alessio Fanelli [00:42:06]: Um, any other interesting research ideas that you've seen, or like maybe things that you cannot pursue a Google that you would be interested in seeing researchers take a step at, I guess you have a lot of researchers. Yeah, I guess you have enough, but our, our research.Jeff Dean [00:42:21]: Our research portfolio is pretty broad. I would say, um, I mean, I think, uh, in terms of research directions, there's a whole bunch of, uh, you know, open problems and how do you make these models reliable and able to do much longer, kind of, uh, more complex tasks that have lots of subtasks. How do you orchestrate, you know, maybe one model that's using other models as tools in order to sort of build, uh, things that can accomplish, uh, you know, much more. Yeah. Significant pieces of work, uh, collectively, then you would ask a single model to do. Um, so that's super interesting. How do you get more verifiable, uh, you know, how do you get RL to work for non-verifiable domains? I think it's a pretty interesting open problem because I think that would broaden out the capabilities of the models, the improvements that you're seeing in both math and coding. Uh, if we could apply those to other less verifiable domains, because we've come up with RL techniques that actually enable us to do that. Uh, effectively, that would, that would really make the models improve quite a lot. I think.Alessio Fanelli [00:43:26]: I'm curious, like when we had Noam Brown on the podcast, he said, um, they already proved you can do it with deep research. Um, you kind of have it with AI mode in a way it's not verifiable. I'm curious if there's any thread that you think is interesting there. Like what is it? Both are like information retrieval of JSON. So I wonder if it's like the retrieval is like the verifiable part. That you can score or what are like, yeah, yeah. How, how would you model that, that problem?Jeff Dean [00:43:55]: Yeah. I mean, I think there are ways of having other models that can evaluate the results of what a first model did, maybe even retrieving. Can you have another model that says, is this things, are these things you retrieved relevant? Or can you rate these 2000 things you retrieved to assess which ones are the 50 most relevant or something? Um, I think those kinds of techniques are actually quite effective. Sometimes I can even be the same model, just prompted differently to be a, you know, a critic as opposed to a, uh, actual retrieval system. Yeah.Shawn Wang [00:44:28]: Um, I do think like there, there is that, that weird cliff where like, it feels like we've done the easy stuff and then now it's, but it always feels like that every year. It's like, oh, like we know, we know, and the next part is super hard and nobody's figured it out. And, uh, exactly with this RLVR thing where like everyone's talking about, well, okay, how do we. the next stage of the non-verifiable stuff. And everyone's like, I don't know, you know, Ellen judge.Jeff Dean [00:44:56]: I mean, I feel like the nice thing about this field is there's lots and lots of smart people thinking about creative solutions to some of the problems that we all see. Uh, because I think everyone sort of sees that the models, you know, are great at some things and they fall down around the edges of those things and, and are not as capable as we'd like in those areas. And then coming up with good techniques and trying those. And seeing which ones actually make a difference is sort of what the whole research aspect of this field is, is pushing forward. And I think that's why it's super interesting. You know, if you think about two years ago, we were struggling with GSM, eight K problems, right? Like, you know, Fred has two rabbits. He gets three more rabbits. How many rabbits does he have? That's a pretty far cry from the kinds of mathematics that the models can, and now you're doing IMO and Erdos problems in pure language. Yeah. Yeah. Pure language. So that is a really, really amazing jump in capabilities in, you know, in a year and a half or something. And I think, um, for other areas, it'd be great if we could make that kind of leap. Uh, and you know, we don't exactly see how to do it for some, some areas, but we do see it for some other areas and we're going to work hard on making that better. Yeah.Shawn Wang [00:46:13]: Yeah.Alessio Fanelli [00:46:14]: Like YouTube thumbnail generation. That would be very helpful. We need that. That would be AGI. We need that.Shawn Wang [00:46:20]: That would be. As far as content creators go.Jeff Dean [00:46:22]: I guess I'm not a YouTube creator, so I don't care that much about that problem, but I guess, uh, many people do.Shawn Wang [00:46:27]: It does. Yeah. It doesn't, it doesn't matter. People do judge books by their covers as it turns out. Um, uh, just to draw a bit on the IMO goal. Um, I'm still not over the fact that a year ago we had alpha proof and alpha geometry and all those things. And then this year we were like, screw that we'll just chuck it into Gemini. Yeah. What's your reflection? Like, I think this, this question about. Like the merger of like symbolic systems and like, and, and LMS, uh, was a very much core belief. And then somewhere along the line, people would just said, Nope, we'll just all do it in the LLM.Jeff Dean [00:47:02]: Yeah. I mean, I think it makes a lot of sense to me because, you know, humans manipulate symbols, but we probably don't have like a symbolic representation in our heads. Right. We have some distributed representation that is neural net, like in some way of lots of different neurons. And activation patterns firing when we see certain things and that enables us to reason and plan and, you know, do chains of thought and, you know, roll them back now that, that approach for solving the problem doesn't seem like it's going to work. I'm going to try this one. And, you know, in a lot of ways we're emulating what we intuitively think, uh, is happening inside real brains in neural net based models. So it never made sense to me to have like completely separate. Uh, discrete, uh, symbolic things, and then a completely different way of, of, uh, you know, thinking about those things.Shawn Wang [00:47:59]: Interesting. Yeah. Uh, I mean, it's maybe seems obvious to you, but it wasn't obvious to me a year ago. Yeah.Jeff Dean [00:48:06]: I mean, I do think like that IMO with, you know, translating to lean and using lean and then the next year and also a specialized geometry model. And then this year switching to a single unified model. That is roughly the production model with a little bit more inference budget, uh, is actually, you know, quite good because it shows you that the capabilities of that general model have improved dramatically and, and now you don't need the specialized model. This is actually sort of very similar to the 2013 to 16 era of machine learning, right? Like it used to be, people would train separate models for lots of different, each different problem, right? I have, I want to recognize street signs and something. So I train a street sign. Recognition recognition model, or I want to, you know, decode speech recognition. I have a speech model, right? I think now the era of unified models that do everything is really upon us. And the question is how well do those models generalize to new things they've never been asked to do and they're getting better and better.Shawn Wang [00:49:10]: And you don't need domain experts. Like one of my, uh, so I interviewed ETA who was on, who was on that team. Uh, and he was like, yeah, I, I don't know how they work. I don't know where the IMO competition was held. I don't know the rules of it. I just trained the models, the training models. Yeah. Yeah. And it's kind of interesting that like people with these, this like universal skill set of just like machine learning, you just give them data and give them enough compute and they can kind of tackle any task, which is the bitter lesson, I guess. I don't know. Yeah.Jeff Dean [00:49:39]: I mean, I think, uh, general models, uh, will win out over specialized ones in most cases.Shawn Wang [00:49:45]: Uh, so I want to push there a bit. I think there's one hole here, which is like, uh. There's this concept of like, uh, maybe capacity of a model, like abstractly a model can only contain the number of bits that it has. And, uh, and so it, you know, God knows like Gemini pro is like one to 10 trillion parameters. We don't know, but, uh, the Gemma models, for example, right? Like a lot of people want like the open source local models that are like that, that, that, and, and, uh, they have some knowledge, which is not necessary, right? Like they can't know everything like, like you have the. The luxury of you have the big model and big model should be able to capable of everything. But like when, when you're distilling and you're going down to the small models, you know, you're actually memorizing things that are not useful. Yeah. And so like, how do we, I guess, do we want to extract that? Can we, can we divorce knowledge from reasoning, you know?Jeff Dean [00:50:38]: Yeah. I mean, I think you do want the model to be most effective at reasoning if it can retrieve things, right? Because having the model devote precious parameter space. To remembering obscure facts that could be looked up is actually not the best use of that parameter space, right? Like you might prefer something that is more generally useful in more settings than this obscure fact that it has. Um, so I think that's always attention at the same time. You also don't want your model to be kind of completely detached from, you know, knowing stuff about the world, right? Like it's probably useful to know how long the golden gate be. Bridges just as a general sense of like how long are bridges, right? And, uh, it should have that kind of knowledge. It maybe doesn't need to know how long some teeny little bridge in some other more obscure part of the world is, but, uh, it does help it to have a fair bit of world knowledge and the bigger your model is, the more you can have. Uh, but I do think combining retrieval with sort of reasoning and making the model really good at doing multiple stages of retrieval. Yeah.Shawn Wang [00:51:49]: And reasoning through the intermediate retrieval results is going to be a, a pretty effective way of making the model seem much more capable, because if you think about, say, a personal Gemini, yeah, right?Jeff Dean [00:52:01]: Like we're not going to train Gemini on my email. Probably we'd rather have a single model that, uh, we can then use and use being able to retrieve from my email as a tool and have the model reason about it and retrieve from my photos or whatever, uh, and then make use of that and have multiple. Um, you know, uh, stages of interaction. that makes sense.Alessio Fanelli [00:52:24]: Do you think the vertical models are like, uh, interesting pursuit? Like when people are like, oh, we're building the best healthcare LLM, we're building the best law LLM, are those kind of like short-term stopgaps or?Jeff Dean [00:52:37]: No, I mean, I think, I think vertical models are interesting. Like you want them to start from a pretty good base model, but then you can sort of, uh, sort of viewing them, view them as enriching the data. Data distribution for that particular vertical domain for healthcare, say, um, we're probably not going to train or for say robotics. We're probably not going to train Gemini on all possible robotics data. We, you could train it on because we want it to have a balanced set of capabilities. Um, so we'll expose it to some robotics data, but if you're trying to build a really, really good robotics model, you're going to want to start with that and then train it on more robotics data. And then maybe that would. It's multilingual translation capability, but improve its robotics capabilities. And we're always making these kind of, uh, you know, trade-offs in the data mix that we train the base Gemini models on. You know, we'd love to include data from 200 more languages and as much data as we have for those languages, but that's going to displace some other capabilities of the model. It won't be as good at, um, you know, Pearl programming, you know, it'll still be good at Python programming. Cause we'll include it. Enough. Of that, but there's other long tail computer languages or coding capabilities that it may suffer on or multi, uh, multimodal reasoning capabilities may suffer. Cause we didn't get to expose it to as much data there, but it's really good at multilingual things. So I, I think some combination of specialized models, maybe more modular models. So it'd be nice to have the capability to have those 200 languages, plus this awesome robotics model, plus this awesome healthcare, uh, module that all can be knitted together to work in concert and called upon in different circumstances. Right? Like if I have a health related thing, then it should enable using this health module in conjunction with the main base model to be even better at those kinds of things. Yeah.Shawn Wang [00:54:36]: Installable knowledge. Yeah.Jeff Dean [00:54:37]: Right.Shawn Wang [00:54:38]: Just download as a, as a package.Jeff Dean [00:54:39]: And some of that installable stuff can come from retrieval, but some of it probably should come from preloaded training on, you know, uh, a hundred billion tokens or a trillion tokens of health data. Yeah.Shawn Wang [00:54:51]: And for listeners, I think, uh, I will highlight the Gemma three end paper where they, there was a little bit of that, I think. Yeah.Alessio Fanelli [00:54:56]: Yeah. I guess the question is like, how many billions of tokens do you need to outpace the frontier model improvements? You know, it's like, if I have to make this model better healthcare and the main. Gemini model is still improving. Do I need 50 billion tokens? Can I do it with a hundred, if I need a trillion healthcare tokens, it's like, they're probably not out there that you don't have, you know, I think that's really like the.Jeff Dean [00:55:21]: Well, I mean, I think healthcare is a particularly challenging domain, so there's a lot of healthcare data that, you know, we don't have access to appropriately, but there's a lot of, you know, uh, healthcare organizations that want to train models on their own data. That is not public healthcare data, uh, not public health. But public healthcare data. Um, so I think there are opportunities there to say, partner with a large healthcare organization and train models for their use that are going to be, you know, more bespoke, but probably, uh, might be better than a general model trained on say, public data. Yeah.Shawn Wang [00:55:58]: Yeah. I, I believe, uh, by the way, also this is like somewhat related to the language conversation. Uh, I think one of your, your favorite examples was you can put a low resource language in the context and it just learns. Yeah.Jeff Dean [00:56:09]: Oh, yeah, I think the example we used was Calamon, which is truly low resource because it's only spoken by, I think 120 people in the world and there's no written text.Shawn Wang [00:56:20]: So, yeah. So you can just do it that way. Just put it in the context. Yeah. Yeah. But I think your whole data set in the context, right.Jeff Dean [00:56:27]: If you, if you take a language like, uh, you know, Somali or something, there is a fair bit of Somali text in the world that, uh, or Ethiopian Amharic or something, um, you know, we probably. Yeah. Are not putting all the data from those languages into the Gemini based training. We put some of it, but if you put more of it, you'll improve the capabilities of those models.Shawn Wang [00:56:49]: Yeah.Jeff Dean [00:56:49]:

Pakistan's Khawaja Asif said,'US used & dumped us like toilet paper'. He was speaking about Pakistan's terror problem and the country's involvement in Afghanistan conflict. #CutTheClutter Episode 1795 looks at Pakistani Defence Minister's comments in detail, and the country's Afghanistan dilemma, in the backdrop of escalating tensions. ThePrint Editor-In-Chief Shekhar Gupta also takes you back to 2011, when he had explained why India should stay away from Afghanistan after US withdrawal, and leave it to Pakistan.

In episode 2005, Jack and Miles are joined by host of There Are No Girls on the Internet, Bridget Todd, to discuss… Difference Between Europe and The US, The Best We Can Do In The US Is A MAGA Senator Saying NOW I SEE WHAT THE BIG DEAL IS, “Penisgate” Is The Latest Olympics Cheating Scandal and more! The Epstein scandal is taking down Europe’s political class. In the US, they’re getting a pass. The Best We Can Do In The US Is A MAGA Senator Saying NOW I SEE WHAT THE BIG DEAL IS What US ski jumpers think about ‘wild’ penis-gate scandal at 2026 Winter Olympics Who is Anthony Ammirati? Meet the French pole vaulter whose ‘bulge’ cost him a medal Ski jumpers sceptical of penis injection reports Skisprung-Verband reagiert auf Penis-Wirbel Rumors Fly Claiming Olympic Ski Jumpers Are Injecting Their Genitals LISTEN: Slipping Into Darkness by The FunkeesSee omnystudio.com/listener for privacy information.

The Justice Department has released its final tranche of the Epstein files - we'll look at what they have and haven't included. And, the U.S. in a partial government shutdown again, although this is one is expected to be shorter than the record-breaking shutdown that happened during the fall. Plus, another winter storm is hitting the U-S this weekend, this time, hitting parts of the Southeast.Learn more about sponsor message choices: podcastchoices.com/adchoicesNPR Privacy Policy

In episode 1997, Jack and Miles are joined by comedian, actor, writer, and host of Worse Than You, Mo Fry Pasic, to discuss… That’s why Teenage Mutant Ninja Goebbels Is In The Hot Seat... Not Under The Bus Yet, TikTok Isn’t The Only Tech Company Backing ICE, Melania Doc Watch and more! Daily Zeitgeist: Our 2000th Episode is Here!!!... Chuck E Cheese - Most Perfect Day (Rap Song) Scoop: Blame game erupts over Trump team's false claim Alex Pretti sought "massacre" Pressure grows on Stephen Miller after Alex Pretti killing but Trump unlikely to cut ties TikTok users say they can’t upload anti-ICE videos. The company blames tech issues TikTok is investigating why some users can't write 'Epstein' in messages The Trump-approved US TikTok is off to a rough start TikTok Says It’s Not Censoring ‘Free Palestine’ Comments. Users See Something Different TikTok now specifically tracks immigration status and gender identity TikTok alternative Skylight soars to 380K+ users after TikTok US deal finalized TikTok's new terms of service spark backlash, but experts say they're an industry standard Meta Is Blocking Links to ICE List on Facebook, Instagram, and Threads How Amazon Powers ICE’s Deportation Machine Apple Took Down These ICE-Tracking Apps. The Developers Aren't Giving Up SCOOP: Apple Made ICE Agents a Protected Class “Melania” Movie Popcorn Bucket Hits ebay for $29.99 — Can Be Purchased Without Actually Seeing the Film Fuming Melania Puts Trump Aides at Risk of Ouster: Wolff Melania Invites Host of Z-List Celebs to Vanity Doc Premiere LISTEN: Danger by The Lijadu Sisters and also check out NUR-D's music here: https://nurdrocks.com/See omnystudio.com/listener for privacy information.