Podcasts about outputs

#865: Neal and Toby talk about how inflation is heating up to the highest pace in three years. Plus, a whole bunch of FIFA World Cup news and how escorts are cashing in on the AI boom over in Silicon Valley. Hit TV shows are taking much longer in between seasons. Why Gen Z and Millennials looove waiting in lines for their trendy food spots. Finally, Rivian finally delivers its R2 model and the first trailer of the much-anticipated ‘The Social Reckoning' drops.  To learn more visit https://www.sage.com/morningbrew Subscribe to Morning Brew Daily for more of the news you need to start your day. Share the show with a friend, and leave us a review on your favorite podcast app. Listen to Morning Brew Daily Here:⁠ ⁠⁠https://www.swap.fm/l/mbd-note⁠⁠⁠  Watch Morning Brew Daily Here:⁠ ⁠⁠https://www.youtube.com/@MorningBrewDailyShow⁠ This is a paid advertisement. Today's episode of the Morning Brew Daily Show is brought to you by Sage — a trusted global provider and leader in accounting, financial, HR, and payroll technology for small and mid-sized businesses. The following commentary reflects general information about Sage and its products. Specific features, capabilities, and availability may vary by product, region, and customer requirements. To find out more, visit sage.com/morningbrew. Paid endorsement. Brokerage services provided by Open to the Public Investing Inc, member FINRA & SIPC. Advisory services by Public Advisors LLC, SEC-registered adviser.  Investing involves risk. Not investment advice. Agentic Brokerage is an AI-powered conversational tool that allows you to enter instructions for a set of self-directed, recurring transactions (your “Agent”) for your account. Outputs from Agentic Brokerage are provided for informational and illustrative purposes only, and should not be considered investment recommendations or advice. Complete disclosures available at public.com/disclosures. See terms of match program at https://public.com/disclosures/matchprogram. Matched funds must remain in your account for at least 5 years. Match rate and other terms are subject to change at any time. Learn more about your ad choices. Visit megaphone.fm/adchoices

This episode gives you the simple tactics that 10x Claude Research ouputs. Here is the link to the companion Substack blog post with the copy and paste master prompt that will 10x the quality of your Claude AI Research outputs: https://tinyurl.com/Companion-Ep1 My full collection of growth hacks, playbooks, and meta prompts lives on my Substack at: https://ClaudeGenius.com

Producing more content faster is not the same as producing content that matters. In this episode of Content Amplified, Adam Haskew, Associate Director of Brand Experience at Redis, makes the case that AI accelerates your outputs but does nothing for your strategy, and that the gap between the two is where "AI slop" gets made. Adam argues the fix is the unglamorous, old-school stuff most teams skip when they are moving fast: kickoff calls, a genuinely complete brief, and human alignment at the very start of a project, before a single word is generated. He explains why a web page is really the same as an ebook when it comes to planning, why skipping alignment creates a "snowball effect" where small problems amplify downstream, and how about an hour and a half of upfront communication removes most of the noise. He also shares how he owns a brand voice review agent at Redis that every piece of content has to pass through before it ships, and why, quoting musician Nick Cave, AI that has never felt hunger or fear still cannot replace a human point of view. If you are shipping more content than ever but learning nothing from it, this conversation gives you the red flags to watch for and a starting point to fix it.About AdamAdam Haskew is the Associate Director of Brand Experience at Redis, where he leads a three-person team focused on brand voice consistency and accurate messaging across the website, print collateral, and trade show materials. He studied English literature and started his career in magazine publishing in Chattanooga, Tennessee, then worked at software companies, an insurance provider, and SaaS companies in the Bay Area before settling into a remote role at Redis. Adam sees AI as a tool in the toolbox, not a replacement for the human judgment that turns content into something worth reading. He believes the best content starts with a clear brief and human communication, then uses AI to execute against that strategy, never the other way around.Show Notes- Connect with Adam on LinkedIn: https://www.linkedin.com/in/adamhaskew/Text us what you think about this episode!

In this week's throwback episode that originally aired as Episode 222, I am breaking down my Top Lessons from the Top 5 All-Time Episodes of The 20% Podcast. In this week's episode, I took the top lesson from each guest, and will be sharing it on today's episode. Here are the Top 5 Episodes by Listens for The 20% Podcast:5. Nick Cegelski: Get To The Truth4. Erik McKee: Building SaaSBros In Public3. Jen Allen-Knuth: Creating The Evangelist Role2. Ian Koniak: Focus on Outputs, Not Outcomes1. Anthony Natoli - How SaaS Saved His LifeThe Top Lessons Include:Discipline Building in publicGiving more than you receiveDoing more than your job titleFocusing on your outputsControlling what you can controlThese are some of the hardest working people that I know. Many of which have overcome a significant amount of adversity to get to where they are today. These are all incredible humans who are all willing to go above and beyond for their clients, and all 5 guests give more than they receive.Thank you so much for your support If there are any guests you'd like to hear me talk with on The 20% Podcast, send me a message on LinkedIn. Please enjoy this week's episode of The 20% Podcast.____________________________________________________________________________I am now in the early stages of writing my first book! In this book, I will be telling my story of getting into sales and the lessons I have learned so far, and intertwine stories, tips, and advice from the Top Sales Professionals In The World! As a first time author, I want to share these interviews with you all, and take you on this book writing journey with me! Like the show? Subscribe to the email: https://mailchi.mp/a71e58dacffb/welcome-to-the-20-podcast-communityI want your feedback!Reach out to 20percentpodcastquestions@gmail.com, or find me on LinkedIn.

Welcome to the Together We Rise podcast! This is a place where every woman has a seat at the table – and where we affirm, equip, and amplify women's voices globally.In this episode, hosts Cheryl Nembhard and Aubrey Sampson welcome Bette Dickinson, a prophetic artist, author, and speaker known for her unique approach to spiritual life through visual parables. The conversation delves into Betty's journey of discovering her calling, which involves creating art that resonates with spiritual truths. She emphasizes the importance of understanding the health of the vine rather than just focusing on the fruit, a lesson she learned from her experiences with a local vine dresser. This perspective challenges the productivity-obsessed culture that often leads to burnout and neglect of spiritual well-being. The discussion touches on the significance of dormancy in spiritual growth, the necessity of rest, and the importance of surrendering control to God. Bette DickinsonBette Dickinson is a prophetic artist, author, and speaker who creates what she calls “visual parables of the spiritual life,” inviting people into deeper encounters with God through beauty and wonder.She is the author of The Art of Vinemaking and Making Room in Advent, which blend biblical insight with original illustrations and contemplative practices. Through her ministry, Awakening the Soul, Bette creates multi-sensory resources and retreat experiences that help people walk through seasons of life, death, and resurrection while rediscovering the sacred rhythm of abiding in Christ. Find out more at bettedickinson.com.WSC Women to Watch: Kristi BramlettKristi Bramlett has spent years cultivating a multifaceted career as a professional actress, college professor, communications coach, Certified Movement Analyst, and author of The Speaker's Trifecta. With over 30 years of experience helping communicators speak with authenticity and presence, her work centers on guiding speakers to communicate with their whole self, mind, body, and soul. Find out more at kristibramlett.com. 

Jedes zweite Tool bekommt gerade einen KI-Chat. Und ausgerechnet der Head of ChatGPT findet das problematisch. Nick Turley stellte auf der OMR 2026 die provokante These auf, dass Chat-Interfaces grundsätzlich an ihre Grenzen stoßen. Sie erinnern ihn an MS-DOS. Damit trifft er, was viele in der Branche gerade denken: Die Welle von KI-Chats ist ein notwendiger Evolutionsschritt, aber noch nicht das Ziel für Amazon PPC.Yarin und Ines diskutieren, warum die nächste Generation von KI-Tools von reaktiv auf proaktiv umschalten muss, was das für Amazon Advertising und PPC-Manager bedeutet und worauf ihr bei der Tool-Auswahl jetzt achten solltet.Alle Themen der Episode im Überblick: Yarin & Ines stellen sich vor (00:00)Warum jeder Anbieter gerade KI-Chats baut (01:40)Nick Turleys These zu Chat-Interfaces (03:36)Die MS-DOS-Analogie: Warum Chat ein Rückschritt sein kann (04:06)Wo Chatbots bei 500 Produkten zusammenbrechen (10:11)Wenn KI ohne Kontrolle handelt: OpenClaw als Beispiel (14:06)Reaktiv vs. proaktiv: Der entscheidende Unterschied (15:09)Determinismus: Warum gleiche Inputs gleiche Outputs brauchen (21:18)Was die Aufgabe des PPC-Managers bleibt (22:38)Worauf ihr bei KI-Tools jetzt achten solltet (25:20)Links & Ressourcen:CORTUA Warteliste: Sichere dir deinen Platz – vor dem offiziellen StartOMR 2026 Talk: Nick Turley (Head of ChatGPT bei OpenAI) im InterviewYarin auf LinkedInInes Ehrhorn auf LinkedInFragen & Anregungen:Hintergründe sowie weiterführende Informationen zum Podcast findest du unter: https://www.adference.com/podcast-vitamin-aFür Fragen und Feedback schreib uns auf LinkedIn: https://www.linkedin.com/in/anna-waag/ oder hinterlasse einen Kommentar auf YouTube: https://www.youtube.com/@ADFERENCEMail: vitamin-a@adference.com

Fault Tolerance for Quantum Inputs and Outputs with Matthias ChristandlWhy This Episode MattersMost discussions of fault tolerance quietly assume a classical-in, classical-out picture: you feed in bits, the noisy quantum machine does its work, and a stable classical answer comes out the other side. Christandl — a mathematically trained quantum information theorist who also leads a Novo Nordisk Foundation–funded life sciences center — argues that this framing is too narrow for the era we are actually entering, where multi-core processors, networked QPUs, and quantum communication links all need to exchange quantum information between noisy machines.If you care about how quantum networks, distributed quantum computers, and quantum simulation workflows for chemistry and biology actually get built, this episode lays out a foundational way of thinking about the problem and connects it directly to current hardware and algorithm co-design.SponsorThis episode is brought to you by Outshift, Cisco's incubation engine. The need for computational power is rapidly increasing in every sector. From drug discovery to material innovation to complex financial modeling, classical systems are reaching their absolute limits. It's time for a paradigm shift. The answer is a scalable quantum network, built on open standards and vendor-agnostic architecture. By uniting distributed quantum devices, you unlock limitless computational power. Learn more about the Cisco Universal Quantum Switch at Outshift.com.Go deeper with the blog post.What We Get IntoWhy the fault tolerance theorem as usually stated leaves out the case that matters most for networking: quantum inputs and quantum outputs.How Christandl's group shows you can still prepare arbitrarily complex quantum states on a noisy machine, paying only one final layer of physical noise rather than collapsing the whole computation.What this means for restoring meaning to quantum channel capacity results in the presence of noisy encoders and decoders.Why distributed quantum computing — multi-core QPUs talking to each other in quantum, not classical, information — is the natural setting for this work.How recent quantum LDPC code work fits in, and why the team is now focused on making encoders and decoders more space-efficient.Christandl's debate with Gil Kalai: which skeptical assumptions are worth taking seriously, and which he thinks the fault tolerance machinery is robust against.The Quantum for Life workflow: zooming in on the quantum-relevant region of a protein–ligand interaction, running a small quantum simulation, and feeding the result into a classical machine-learning pipeline that needs many such small computations.Why "co-design" has replaced "bridging the gap" as the right metaphor for where quantum hardware and quantum software meet.How quantum sensing — for example, magnetic-field sensing with atomic clouds — could one day deliver genuine quantum inputs into a fault-tolerant quantum computer.Resources & LinksGuest LinksMatthias Christandl — University of Copenhagen Research Portal — Official institutional profile with publications and affiliations.Quantum for Life Center — University of Copenhagen — The Novo Nordisk Foundation–funded center Christandl leads, focused on quantum algorithms for the life sciences.UCPH Quantum Hub launch — The cross-faculty quantum community Christandl helped found at the University of Copenhagen.Christandl appointed 2024 Turing Chair — CWI/QuSoft — Background on his honorary visiting chair at QuSoft and CWI in Amsterdam.Papers & ArticlesFault-Tolerant Coding for Quantum Communication (arXiv:2009.07161) — The foundational paper (IEEE TIT 2024, with Müller-Hermes) that motivates the episode: channel coding when the encoder and decoder circuits themselves are noisy.Fundamental Limit on the Power of Entanglement Assistance in Quantum Communication (arXiv:2408.17290) — Christandl and collaborators settle a 2002 conjecture of Bennett et al. on entanglement-assisted capacity (PRL 2025).Asymptotic tensor rank is characterized by polynomials (arXiv:2411.15789) — STOC 2025 result connecting tensor theory to the matrix multiplication exponent.How to Use Quantum Computers for Biomolecular Free Energies (2026)More Quantum Chemistry with Fewer Qubits — Physical Review Research (2024) — The Quantum for Life paper underlying the protein–ligand workflow discussed in the episode.A Cornerstone of Entanglement Theory Restored — Nature Physics (2025) — Christandl's News & Views on the re-proof of the generalized quantum Stein's lemma.Quantum Duel: Matthias Christandl x Gil Kalai Key Quotes & InsightsOn reframing fault tolerance: Christandl argues that the fault tolerance theorem, as usually stated, assumes classical inputs and outputs — but the most important near-term use cases, from networked QPUs to multi-core processors, need quantum inputs and quantum outputs.On the unavoidable final layer of noise: "There will always be a final layer of noise being applied" when a noisy machine prepares a quantum state — and that single layer, not the whole computation, is the real price you pay.On the new metaphor: "A few years back, I would have told you the really important thing is bridging the gap between the hardware and the software. Now it's not anymore about bridging the gap. It's about working together."On Kalai's skepticism: Christandl finds the debate clarifying rather than threatening — the fault tolerance techniques look robust to the noise-model perturbations skeptics raise, and the engineering question is which code, not whether codes work at all.On what quantum advantage in life sciences might actually look like: Not one heroic simulation, but many small, exact quantum computations feeding training data into a much larger classical machine-learning workflow that predicts protein–ligand interactions.Related Episodes

Most AI conversations focus on models. The better conversation focuses on systems. In this episode, we continue our interview with Matt Levenhagen, exploring a practical challenge many developers are facing: integrating AI into business operations without creating costly chaos. The answer is not buying more AI tools. The answer is building an intentional AI Workflow Architecture. About Matt Levenhagen Matt is the founder and CEO of Unified Web Design, a web development agency focused on custom solutions, WordPress development, e-commerce, memberships, and business systems. His background as both a builder and agency owner gave him a unique perspective on where AI creates real leverage instead of superficial automation. Follow Matt on LinkedIn. AI Workflow Architecture Starts with Context Control One of the most important operational realities Matt discussed was token usage. Businesses rushing into AI often underestimate cost scaling. Every interaction with large models consumes resources, and poorly managed context windows dramatically increase operational expenses. Instead of treating AI like unlimited compute, Matt focused on controlling context intentionally. That included: Monitoring token usage Limiting unnecessary memory loading Structuring retrieval systems Using different models for different tasks Preventing oversized prompts This is a systems-thinking problem, not merely a coding problem. Developers who ignore architecture end up with bloated workflows that become financially unsustainable. The fastest way to make AI unprofitable is to send unnecessary context into every request. Why Retrieval Matters More Than Raw Memory A major breakthrough Matt discussed was implementing Retrieval-Augmented Generation (RAG). This matters because AI systems do not need all the information all the time. They need the right information at the right moment. That distinction completely changes system design. Without retrieval architecture: Costs increase Performance slows Outputs become less accurate Hallucinations increase Operational complexity grows RAG allows systems to retrieve semantically relevant information instead of dumping entire databases into prompts. This transforms AI from brute-force processing into intelligent retrieval. The future of AI operations will likely depend less on giant models and more on efficient information orchestration. AI Workflow Architecture Requires Layer Separation Another valuable concept from the conversation involved separating operational layers. Matt described balancing: Local storage Business memory External AI APIs Workflow automation SaaS integrations This layered architecture creates flexibility. Instead of locking the business into one AI provider, workflows remain adaptable. Different models can handle different workloads depending on cost, complexity, and accuracy requirements. This becomes increasingly important as pricing models fluctuate. Businesses relying entirely on one provider risk operational instability if pricing changes dramatically. Layer separation reduces that risk. The businesses that survive AI cost volatility will be the ones architected for flexibility instead of dependency. Why Embedded AI Features Often Disappoint Matt also discussed the growing wave of SaaS AI integrations. Every platform now markets AI capabilities: Project management tools Communication platforms CRM systems Design software Documentation systems Yet many users feel underwhelmed. The reason is architectural isolation. These tools only understand limited slices of operational context. They automate micro-tasks but rarely improve larger workflows. That creates a false impression that AI itself lacks value when the real issue is fragmented systems. AI becomes more useful as the organizational context becomes more connected. This is why developers building custom operational layers still maintain an enormous strategic advantage. AI Workflow Architecture Is an Operational Discipline The strongest insight from these episodes may be that AI implementation is becoming operational engineering. Success now depends on: Information structure Retrieval design Workflow sequencing Context prioritization Cost management Human oversight This moves AI away from novelty experimentation and toward infrastructure planning. Businesses that treat AI casually will likely accumulate technical debt quickly. Businesses that approach AI architecturally will build scalable operational leverage. AI is no longer just a development tool. It is becoming an operational systems discipline. Developers Must Learn Economic Thinking One overlooked topic in AI discussions is economics. Matt repeatedly referenced balancing capability with cost. This becomes critical because AI pricing models are still evolving rapidly. Businesses that ignore usage economics may accidentally build systems that become financially impossible to scale. Developers now need to think beyond: Can this be built? They also need to ask: Can this be sustained? Can this scale economically? Can context costs remain controlled? Can cheaper models handle simpler tasks? This represents a major evolution in modern software architecture. Review your current AI workflows and identify where unnecessary context or oversized prompts may be increasing costs. Conclusion AI Workflow Architecture is rapidly becoming one of the most important technical disciplines for modern developers. Matt Levenhagen's approach demonstrates that successful AI implementation is less about chasing the newest model and more about designing sustainable operational systems. The companies that gain long-term advantage from AI will not necessarily be the companies using the largest models. They will be the companies with the best architecture. Stay Connected: Join the Developreneur Community

Spy Academy - https://www.marketingsharks.com/spy-academy-powered-by-activity-book-generator/I forgot to mention the demo video, click the link to my review at marketingsharks.com then click the product boxes to get to the sales page, scroll down a wee bit, Demo Video! It looks to be an AI introduction, but push past that, from the 4 minute mark to the 8 or 9 minute mark, Amber herself shows how it all works!Here's what the Spy Mission Books Builder does:✓ Generates a full spy narrative with characters, objectives, and plot twists✓ Weaves real cipher puzzles and code challenges into the story✓ Three difficulty levels — from junior recruit to elite field agent✓ Outputs as a ready-to-publish PDF✓ Every book is unique. Click generate, get a completely original spy mission. Click again — a different one.

(Disclaimer: erstellt mit ChatGPT)Hallo liebe Community,

Natalie Marcotullio & Mark Kilens invited me on their podcast GTM News Desk.We discuss:How do you build messaging that actually sticks across every team that touches it? How do you know when an "insight" is genuinely worth acting on or when it's just a pattern dressed up in fancy words? What does it take to teach AI to think like you, not just write like one version of you?Mojo Founder Eric Holland and I reveal what's really broken about how B2B companies develop and distribute their messaging. We dig into why most customer research is pulled from the most biased data available and how to document your decision-making, not just your knowledge, so AI can actually replicate your judgment. We also provide answers to why chasing every new AI workflow is making most marketers worse, not better. Eric and I challenge the idea that volume of insights equals quality of insights, arguing that a single finding that changes how your whole company goes to market is worth more than a hundred tagged Gong quotes.Jump in:(00:00) We need a higher standard for b2b research(00:35) What CMOs are actually talking about at conferences right now(04:40) How shame and fear became the default AI adoption strategy(08:41) The difference between AI-exhausted and AI-excited teams(12:44) Four Tendencies and what it means for AI adoption(16:31) Why B2B messaging is broken even at well-funded startups(19:14) What separates Mojo from just pointing Claude at your Notion(27:26) What good writing actually feels like and whether AI can get there(32:17) The solar plexus test and how to know when your copy is working(34:05) Why nuance and emotion are the real gap in AI writing(39:27) Why most insight tools are just pattern-matching machines(44:19) How to check if your researcher is actually any good(47:30) The danger of everyone using the same workflows to find the same "insights"(50:01) What B2B marketing gets wrong that B2C figured out long ago(53:21) Go deep on one thing for a quarter(55:22) Why you should only build what you already know in your sleepConnect with Eric: https://www.linkedin.com/in/eric-holland-not-a-marketer/ Connect with Anna: https://www.linkedin.com/in/annafurmanov/ Mojo: https://mojopmm.com/Mojo + Moxie: https://mojo-and-moxie.com/Natalie: https://www.linkedin.com/in/natalie-marcotullio/Mark: https://www.linkedin.com/in/markkilens/

Schickt uns euer Feedback zur EpisodeDas Internet fühlt sich plötzlich fremd an: zu glatt, zu voll, zu laut und manchmal so künstlich, dass wir selbst echten Menschen nicht mehr trauen. Wir nehmen die Dead Internet Theory ernst, ohne in Verschwörung abzurutschen, und schauen stattdessen auf die Resultate: KI-Slop, Content-Farmen, Bots, Propaganda und eine Aufmerksamkeitsökonomie, die Masse belohnt statt Sinn. Wenn Kommunikation ohne echte Intention entsteht, kippt der Diskursraum und es wird immer schwerer, „gute“ Informationen zwischen den Klatschblättern zu finden.Gleichzeitig verlagert sich Vertrauen ausgerechnet zu LLMs wie ChatGPT, Claude oder Perplexity, weil sie den Informationsdschungel scheinbar filtern. Wir sprechen darüber, warum das LLM damit zum Gatekeeper wird, wie sich SEO Richtung GEO verschiebt, und weshalb der Wettbewerb der KI-Firmen strukturell zu Monopolen drängt. Dual-Use, Government-Contracts und „too big to fail“ sind nicht nur Schlagworte, sondern echte Pfade, über die KI-Infrastruktur staatstragend werden kann. Hinter dem Hype liegt zudem viel menschliche Arbeit: RLHF, Clickworker und eine „polierte“ Oberfläche, die Desinformation genauso glattziehen kann wie hilfreiches Wissen.Wir bleiben nicht im Doom hängen, sondern suchen konkrete Gegenentwürfe: Transparenz, Open Source, Datensouveränität sowie EU-Regeln wie Digital Services Act und Digital Markets Act. Und wir zeigen, wie KI auch anders wirken kann, etwa als Moderations- und Auswertungswerkzeug für bessere Diskussionen, inklusive unserer kleinen Habermas-Maschine im Unternehmensalltag. Zum Schluss wird es praktisch: Welche neuen Kompetenzen brauchen wir, um Blackbox-Autorität zu widerstehen, Outputs einzuordnen und Ziele sauber zu setzen, ohne Lernarbeit und Urteilskraft zu verlieren?

The downside of powerful, autonomous models that can think and act?

Warren Buffett once said it's only when the tide goes out that you discover who's been swimming naked. This week, the tide went out on several fronts simultaneously, and what it revealed was uncomfortable, instructive, and in some cases, long overdue.France opened the week with a breach that should trouble every government running centralised identity infrastructure. Up to 19 million records tied to passports, ID cards, and driver's licenses are now circulating on criminal forums. What makes this worse than a typical data leak is the context: a similar dataset from the same agency surfaced in 2025. This wasn't a surprise attack on a hardened target. It was a recurring failure wearing the face of a solved problem.The Bitwarden supply chain story carried a similar energy. No vaults were cracked, no passwords were stolen, and most users never noticed a thing. But a malicious package briefly moved through npm as part of the Checkmarx campaign, targeting the developers who build the software everyone else depends on. The lesson isn't technical — it's structural. Your security posture now extends to every build pipeline, every dependency, and every automation script upstream of your product.Then came FAST16.SYS, and the week shifted into something darker. This rootkit, which appears to predate Stuxnet, didn't steal data or trigger alarms. It quietly altered precision calculations in memory while leaving every file on disk untouched. Systems looked healthy. Outputs looked reasonable. The only thing wrong was the answer. It is the most patient form of sabotage imaginable, and it reframes what advanced threats are actually capable of when detection, not damage, is the real objective.AI brought its own escalation this week. Researchers are now using AI systems to attack other AI systems at machine speed — probing, learning, and refining exploits far faster than any human team. At the same time, agent browsers like Interceptor are quietly repositioning the browser itself as an autonomous actor, raising legitimate questions about oversight when software is doing the clicking, typing, and deciding on your behalf.Anthropic's Mythos model access story tied several threads together neatly. Contractor credentials, open-source reconnaissance, and data exposed in a third-party breach combined to give a small group access to a restricted model. The intent was curiosity, not sabotage — but the mechanism was a textbook illustration of how third-party access chains create exposure that principal organisations rarely see coming.Apple closed out the privacy section with a rare win, patching a logging bug that had been silently retaining Signal message fragments for up to a month — long after deletion, long after the app was removed. The FBI had already used it in court. The patch is clean and the fix is automatic, but the episode is a pointed reminder that ephemeral and permanent are closer together than most people assume.The week closed on strategy. OpenAI and Microsoft have restructured their foundational partnership, removing exclusivity and capping revenue payments. The AI infrastructure layer is becoming contested ground, and this deal confirms that no single partnership, however dominant it once appeared, is permanent.This week's stories didn't shout. They accumulated. And that, more than anything, is the point.

„Wer KI nutzt, ohne zu entscheiden, nach welchen Regeln sie arbeitet, entscheidet trotzdem. Nur eben nicht bewusst."Eine Episode über Wildwuchs, Gedankenlosigkeit und die Frage, wer eigentlich die Verantwortung trägt für das, was KI in die Welt bringt.Die Rubrik #ConsciousIntelligence ist meine Solo-Reihe im Podcast. Hier denke ich laut: über Business, Kommunikation und Verantwortung, über KI, Marketing und Führung, immer mit dem Blickwinkel auf Bewusstsein und Ethik.Wir reden tagtäglich über Prompts, Tools und Automatisierung. Aber die Frage, die kaum jemand stellt: Wer trägt eigentlich die Verantwortung für das, was KI Tag für Tag in die Welt bringt?Für mich gilt: KI-Nutzung ohne Verantwortung ist keine Privatsache. Denn was wir KI übergeben, was wir mit ihr produzieren und wie wir diese Outputs in die Welt bringen, hat Auswirkungen: Auf Inhalte, auf Kommunikation und auf das, was wir kollektiv als Standard akzeptieren.In dieser Folge erfährst du:warum KI-Nutzung keine Privatsache ist und welche Auswirkungen sie auf Inhalte, Kommunikation und gesellschaftliche Standards hatwelche drei Muster ich in der Praxis immer wieder beobachtewas ein verantwortungsvolles Setup konkret bedeutet und warum es nicht beim Prompting beginnt, sondern davor undwarum verantwortungsvolle KI-Nutzung keine Frage des guten Willens ist, sondern eine Frage deiner Systemarchitektur.Wenn du KI bereits nutzt, aber das Gefühl nicht loswirst, dass die Ergebnisse irgendwie nicht zu dir passen, dann ist diese Folge genau der richtige Einstieg.Mehr über mich und meine Arbeit findest du hier:

HEARD IN THIS EPISODE ◼ Why the four-year decline in staffing revenue isn't a market problem — it's a sales behavior problem that leaders are actively making worse by obsessing over the wrong metrics ◼ How COVID didn't just change where salespeople work — it quietly made them lazy, and why most sales leaders haven't noticed yet ◼ What the best staffing firms are doing differently right now that almost nobody else is: pulling recruiters into the sales process before the deal is even close ◼ Why your ICP is probably wrong — and how getting more specific with a smaller target list will generate more revenue than casting a wider net ever could ◼ How one simple mental shift — from account management to pre-client acquisition — could completely rewire the way your sales team operates EXPECT TO LEARN This episode will fundamentally challenge how you think about sales activity and what it's actually worth. You'll walk away with a sharper framework for diagnosing what's broken in your sales organization, whether it's your people, your process, or the pressure you're putting on both, and a clear-eyed view of why doing less, more deliberately, consistently beats doing more with less intention. If you lead a staffing firm or a sales team, this conversation will make you uncomfortable in exactly the right way.   KEY MOMENTS [00:01] – The #1 mistake holding firms back [01:08] – Chasing shiny objects kills revenue [02:21] – What to do when business is down [04:08] – How to prioritize your fix-it list [05:28] – Did COVID break the staffing industry? [07:02] – How COVID made staffing sales reps lazy [08:21] – Outputs vs. outcomes in sales [09:18] – Rewarding the wrong KPIs [10:48] – What consultative selling looks like [12:49] – Where firms get their ICP wrong [14:45] – Messaging by ICP level [16:38] – The Under Armour sales analogy [18:58] – Research has never been easier [23:00] – What top firms do differently [24:45] – Speed closes deals, not relationships [27:52] – Fixing staffing's reputation problem [31:42] – Sales reps as pre-client managers [33:26] – Optimistic or pessimistic on 2026? [35:41] – Coach in the moment, not end of month [37:17] – About AFM Strategic Partners [42:20] – Rapid fire: book that changed her life [43:24] – Advice for new staffing professionals ABOUT THE GUEST Anna Frazzetto is the Founder and CEO of AFM Strategic Partners and one of the most recognized voices in staffing and sales leadership — named to Staffing Industry Analysts' Global Power 150 Women in Staffing for seven consecutive years. She built her career scaling businesses from the ground up, including growing a solutions practice from $3M to $100M, and has led global sales transformations across some of the most complex corners of the industry. Her perspective is rare because it spans both sides of the table: the strategic and the deeply human, shaped in part by her experience as a cancer survivor who chose to bet on herself and start something new. Her new book, *Sales Leadership in Action*, distills over 100 field-tested tips for sales leaders who want to stop firefighting and start building teams that close. ABOUT THE HOST Brad Bialy is a trusted voice and highly sought-after speaker in the staffing and recruiting industry, known for helping firms grow through integrated marketing, sales, and recruiting strategies. With over 13 years at Haley Marketing and a proven track record guiding hundreds of firms, Brad brings deep expertise and a fresh, actionable perspective to every engagement. He's the host of Take the Stage and InSights, two of the staffing industry's leading podcasts with more than 225,000 downloads. SPONSORS AND OFFERS Book a 30-minute marketing consultation with host Brad Bialy: https://bit.ly/Bialy30 Benefits in a Card helps staffing firms offer meaningful benefits to their entire workforce through flexible, unbundled plans designed for high-turnover environments—making it easier to control costs, improve retention, and stay competitive. https://www.BenefitsInACard.com TRICOM partners with staffing firms as an asset-based lender and full-service back-office provider, helping owners scale confidently by reducing risk and easing the operational strain of payroll, cash flow, and administration. https://www.tricom.com  

With a record corn crop to move, the corn industry is on the hunt for new and innovative uses for America's crop. And one solution may be found in one of the fastest-growing sectors of the clothing and textiles sector—athleisure wear. The popular clothing style—yoga pants, joggers, hoodies and more—combines high fashion with high-functionality and comfort, and has been gaining in popularity for years. But as with any product that is sourced from petrochemicals, there is an opportunity to replace the oil-based feedstock with one that is sourced from corn. And at Qore, a joint venture between Cargill and HELM, they're working on making this a possibility. So in this episode, we talk to Andrea Vanderhoff, Director of Technology and Sustainability at Qore, to learn more about how their QIRA technology is opening new avenues for corn-based products to penetrate the textiles market, including in athleisure wear. And, NCGA Director of Outputs and Measurements Harley Janssen joins us as well to talk about the potential impacts and benefits for the corn industry. To learn more about Qore and QIRA, visit www.myqira.com

A Brand Strategy is often dismissed as ‘fluff' because it's poorly defined, inconsistently applied, and confused with tactics, communications, advertising, or visual identity.In Part 2 of this series, we move beyond the theory of brand as a function of memory and why buyers buy to define specific strategic outputs, explain how they're informed, and model scenarios in which they're used. This episode presents a unifying idea of brand strategy that holds up for marketers and remains accessible to non-marketers.This episode covers:How brand strategy operates as a set of outputs that shape decisions across an organisation Scenario modelling across two distinct operating environments: a B2B accounting firm and a high-urgency B2C retailer How an effective brand strategy aligns incentives, operating models, and capability, not just communications and advertising outputs Why most competitive advantage comes from structure, not messaging or creativity A practical test to determine whether something is strategic or tacticalThe core outputs of brand strategy, how they're informed, and how they're used The role of memory, association, and expectation in reinforcing or violating brand strategy Brand Strategist Jack Ferguson hosts this episode.Helpful Links:Where to find Jack:- Find Jack on LinkedIn- Find Jack at his WebsiteWhere to find The Push:- Follow The Push on LinkedIn- Subscribe to The Push on YouTube Music- Follow The Push on TikTok- Follow The Push on Instagram- Visit The Push WebsiteResources Mentioned:- RSPCA Commercial- Budweiser + Jay Z Commercial- 2013 Budweiser Super Bowl Ad

The dominant structural shift highlighted is the movement of value from AI-driven features to the ownership and governance of the control plane—specifically, entities that set boundaries, maintain proof, and keep automated workflows within defined limits. This shift is evidenced by workforce polling from Quinnipiac University, business formation trends tracked by the Bank of America Institute and Census Bureau data, and product launches from vendors like TeamViewer and KnowBefore. These developments underscore a growing reliance on automation where traditional human oversight is minimized, and technology increasingly assumes direct control over work execution. The episode details workforce sentiment, citing a Quinnipiac University poll where only 15% of respondents expressed willingness to work for an AI boss, and 70% anticipated AI would reduce job opportunities. Bank of America Institute data notes a 15% year-over-year increase in high propensity businesses—those likely to launch—while businesses planning to hire have fallen by 4%. TeamViewer has introduced TIA Reporting, which generates dashboards via natural language prompts, reducing specialist requirements. KnowBefore's ADA Orchestration automates security awareness scheduling and execution, reportedly shortening setup times from hours to seconds. These examples show how vendors are deploying AI tools that replace specific manual oversight with algorithmic management. Supporting developments reinforce the governance gap. According to a CIO Dive report, 96% of C-suite leaders expect productivity gains from AI, yet 77% of employees report increased workloads, signaling misalignment between leadership intent and actual outcomes. Tech Bullion reveals 60% of organizations have AI integrated in at least one core function, with 65% using generative AI regularly, but fewer than a quarter have operationalized ethical AI frameworks. The Verge covers enhancements to Anthropics' tools that embed guardrails where organizational controls are lacking. Additional survey data from TechCrunch shows that usage of AI is growing while trust in its outputs remains weak; only 24% of respondents trust AI most of the time. Operationally, the implication is clear for MSPs and IT leaders: as organizations reduce human oversight and delegate more work to automation, the auditability, accountability, and control of automated workflows become direct contractual risk. Control layers—such as logging, exception handling, approval thresholds—must be productized and priced, not treated as informal advisory work. Liability for automation failures must be clearly assigned and managed through contractual terms, with automation incident response separated from standard support. Without enforceable governance and evidence of control, MSPs risk absorbing unpaid remediation work as clients expect both automation benefits and assurance of outcome. 00:00 Bossless Workforce 03:22 AI, No Guardrails 05:45 Govern or Absorb 08:41 Why Do We Care?  Supported by:  Nerdio HaloPSA  

Focus on Feedlots: Continued Heavy Cattle NASA STELLA at Ag Tech Day Reducing Corn Silage in Cow's Diet   00:01:05 – Focus on Feedlots: Continued Heavy Cattle: Justin Waggoner, K-State beef cattle specialist, starts today's show as he recaps the recent "Focus on Feedlots" report and where cattle are currently finishing in terms of weight. Focus on Feedlots KSUBeef.org jwaggon@ksu.edu    00:12:05 – NASA STELLA at Ag Tech Day: The show continues with Jacob Orser, program support specialist with NASA Acres, as he discusses NASA's STELLA and what he will be teaching kids at the upcoming Ag Tech Day. NASA - STELLA Ag Tech Day   00:23:05 – Reducing Corn Silage in Cow's Diet: K-State dairy specialist, Mike Brouk, ends the showing saying how recent studies show that a BMR male sterile sorghum hybrid can effectively replace about 25-30% of the corn silage in a lactating cow's diet.       Send comments, questions or requests for copies of past programs to ksrenews@ksu.edu.   Agriculture Today is a daily program featuring Kansas State University agricultural specialists and other experts examining ag issues facing Kansas and the nation. It is hosted by Shelby Varner and distributed to radio stations throughout Kansas and as a daily podcast.   K‑State Extension is a short name for the Kansas State University Cooperative Extension Service, a program designed to generate and distribute useful knowledge for the well‑being of Kansans. Supported by county, state, federal and private funds, the program has county Extension offices statewide. Its headquarters is on the K‑State campus in Manhattan. For more information, visit Extension.ksu.edu. K-State Extension is an equal opportunity provider and employer.

The Joint Readiness Training Center is pleased to present the one-hundredth-and-forty-third episode to air on ‘The Crucible - The JRTC Experience.' Hosted by MAJ David Pfaltzgraff, the BDE Executive Officer Observer-Coach-Trainer and MAJ Marc Howle, the Brigade Senior Engineer / Protection OCT for Brigade Command & Control (BDE HQ), on behalf of the Commander of Ops Group (COG). Today's guests are experts across JRTC: MSG Jared Cawthon as the BDE Fires Support NCOIC, MSG Randell Conway as the BDE Intelligence NCOIC OCT, both from BC2 (BDE HQ), and MAJ Lorenzo Evans is the Support Operations Plans Officer OCT for TF Sustainment (DSSB / LSB).   This episode focuses on the critical outputs of the military decision-making process (MDMP) and how their quality directly determines a unit's ability to execute in combat. Rather than viewing MDMP as a series of steps, the discussion emphasizes that its true value lies in the products it produces—clear commander's guidance, refined mission statements, synchronized warfighting function inputs, and shared fighting products that enable subordinate units to act. Key outputs such as planning guidance, initial and refined timelines, targeting products, and decision support tools are highlighted as essential for translating analysis into executable operations. When done correctly, these outputs create a common understanding across the formation and allow units to operate with speed, clarity, and purpose in a complex environment.   The conversation also underscores that poor or incomplete MDMP outputs are often the root cause of friction during execution. Vague guidance, inconsistent graphics, and lack of version control lead to desynchronized efforts and missed opportunities on the battlefield. Best practices focus on producing simple, clear, and timely outputs that are continuously refined through running estimates and rehearsals. The importance of early dissemination, shared digital and analog products, and enforcing standards across the staff is reinforced to ensure all echelons are aligned. Ultimately, the episode highlights that MDMP is only as effective as the outputs it delivers, and units that master these products gain a decisive advantage in large-scale combat operations.   Part of S13 “Hip Pocket Training” series.   For additional information and insights from this episode, please check-out our Instagram page @the_jrtc_crucible_podcast   Be sure to follow us on social media to keep up with the latest warfighting TTPs learned through the crucible that is the Joint Readiness Training Center.   Follow us by going to: https://linktr.ee/jrtc and then selecting your preferred podcast format.   Again, we'd like to thank our guests for participating. Don't forget to like, subscribe, and review us wherever you listen or watch your podcasts — and be sure to stay tuned for more in the near future.   “The Crucible – The JRTC Experience” is a product of the Joint Readiness Training Center.

In this episode of The Ross Simmonds Show, Ross sits down with Britney Muller, AI educator and founder of Orange Labs, to unpack what marketers are getting wrong about large language models, why reverse engineering ChatGPT is a dead end, and how to build real leverage in a probabilistic world. From practical AI workflows to the ethical risks shaping the future of the industry, this is a first-principles breakdown of what actually matters next. Key Takeaways and Insights: 1. AI is not search, it is a different machine entirely - LLMs are probabilistic word prediction systems, not ranking engines. There are no ranking factors inside ChatGPT and no URLs in its training data. - Most marketers are forcing AI into an outdated SEO mental model, and new technology requires a new framework. 2. Understanding RAG and how visibility actually works - LLMs are often paired with real-time search to stay current, but the core model and the retrieval layer are two separate systems. - Visibility in AI requires influence across both training data and search ecosystems, and SEO still matters even as the mechanics are shifting. 3. Brand mentions over backlinks - LLMs magnify what appears most frequently in training data, which means contextual brand mentions are becoming leverage. - One startup paid for brand mentions on commonly retrieved URLs rather than links and it worked. Distribution across relevant conversations increases the probability of surfacing. 4. Why you cannot reverse engineer LLMs - There is no deterministic ranking system to hack. Outputs vary across identical prompts because of probabilistic modeling. - Most AI tracking tools rely on synthetic prompts and crude metrics. Guarantees in GEO are dangerous and honesty builds trust. 5. Build your own AI tracking stack - Internal tools are now cheaper and more powerful than off-the-shelf platforms. Running prompts multiple times per day allows teams to measure probability ranges. - APIs allow thousands of queries at minimal cost. Control your data and do not outsource your intelligence. 6. Real AI workflows built by marketers - Competitive engagement scraping combined with AI-personalized outreach is producing 80 percent response rates. HARO filtering systems can now auto-draft responses inside Slack in real time. - The common thread across every workflow that works is the same: start with a clear problem, then layer in AI. 7. AI as personal leverage - Brittany used ChatGPT to win a home bidding war with a personalized letter and reframed a payment dispute email as a lawyer, which resulted in payment within 30 minutes. - AI is not just marketing leverage. It is life leverage. Literacy creates power. 8. Is SEO dead? Not quite. - Google patents suggest AI-first interfaces may replace traditional SERPs, and organic traffic levels will likely not return to pre-AI highs. - The pie may shrink but search will not disappear. Off-site distribution and social proof will matter more than ever. 9. The ethical risks of AI power - A small group of decision-makers controls foundational AI systems, and the incentives in place favor hype cycles and growth over accountability. - Reinforcement learning optimizes for pleasing users, not truth. AI literacy must include understanding bias and power structures. 10. The rise of AI agents - Early agents were mostly hype, but new iterations like Claude Chrome integrations can now visually interpret and act inside browsers using screenshot-based reasoning. -The future of marketing may involve AI transacting on behalf of users entirely, and execution changes workflows. Resources & Tools:

In this episode, Morgan sits down with serial entrepreneur, tech founder, and sales obsessive Darren Lee to talk about building multiple seven-figure businesses, why most AI is garbage, and what actually separates the people who scale from those who stay stuck. Darren shares how he lost over €100K hiring 60 people in a year, the brutal truth about A players vs B players, and why the best startups have no money. They also get into running away from pain vs chasing goals, decision-making frameworks for when your business gets bigger, working with your partner every single day, and why learning sales is the only thing that matters.Episode Timestamps0:00 Trailer0:48 Introduction: Meet Darren1:31 What is Aura?4:45 Starting a New Company9:24 Most AI is Garbage15:45 Systems Thinking Background18:23 Inputs, Outputs, Outcomes20:20 Disadvantaged Backgrounds Win24:07 The Economy Flight Story27:18 Someone Dumber is Winning32:51 Lost €100K Hiring 60 People33:47 A Players vs B Players36:17 Building a Hiring Engine48:50 One-Way vs Two-Way Doors53:24 Mentors and Category Kings57:15 Working With Your Partner1:01:02 Ten Year Vision1:03:00 Advice to His 18-Year-Old SelfAbout DarrenDarren Lee is a serial entrepreneur, tech founder, and sales systems expert. He runs multiple seven-figure businesses including a media company, education business, and Aura, an AI-powered sales management platform. With a background in engineering and startups, Darren is obsessed with process design, hiring A players, and building scalable systems. Based in Bali, he works alongside his wife Elise and has built a reputation for helping coaches and agency owners scale through systematic sales processes. This is his first time on the podcast.Connect with Darrenhttps://www.instagram.com/darrenlee.ks/?hl=en Connect with Mehttps://www.youtube.com/@morgantnelsonhttps://www.instagram.com/morgantnelson

Zum Finale der 12. Staffel spreche ich mit Prof.'in Elisabeth Mayweg (Humboldt-Universität zu Berlin) über AI Literacy als Metakompetenz, die sie als Zusammenspiel aus informativem Grundverständnis von KI-Systemen, der Fähigkeit zur kritischen Einordnung von Outputs sowie einer strategischen Nutzungskompetenz definiert. Hieran anknüpfend wirbt Mayweg für ein „kritisch reflektiertes Vertrauen" in generative KI jenseits von Technik-Euphorie und Technik-Skepsis. Wir thematisieren wie Studierende generative KI heute tatsächlich nutzen, warum das geteilte Nicht-Wissen von Lehrenden und Studierenden eine didaktische Chance sein kann, welche Rolle KI beim Denken und Schreiben spielt; und was Lehrende konkret tun können, um AI Literacy in der eigenen Lehre zu verankern.

ChatGPT 5.4 und NotebookLM sind gängige KI-Tools im Jahr 2026 für Trainer, Berater, Coaches Kein Hype. Sondern Praxis. In dieser Episode geht es um meinen direkten Praxistest mit ChatGPT 5.4 und NotebookLM – und um die Frage, was diese Entwicklungen wirklich für Trainer:innen, Berater:innen und Coaches bedeuten. Ich teile, was ich seit dem Release konkret ausprobiert habe, was mich überrascht hat und welche Konsequenzen ich daraus für Recherche, Content-Produktion, Prompt Engineering und Lerndesign ableite. Im Zentrum steht nicht die Frage: Welches Tool ist besser? Entscheidend ist etwas anderes: Welches Tool passt zu welchem Zweck? Genau darin liegt die eigentliche Kompetenz. Außerdem schaue ich auf die neuen Bildstile in NotebookLM, auf ihre mögliche Wirkung für unterschiedliche Zielgruppen und auf die Grenze, an der aus Spielerei wieder strategische Relevanz werden muss. Ein zentrales Fazit der Episode: KI verbessert Lernen nicht automatisch. Sie kann Lernen deutlich unterstützen – wenn sie bewusst eingesetzt wird. Nachhaltiger Lernerfolg braucht weiterhin Reflexion, Transfer und Selbstregulation. In dieser Folge erfährst du:

Are numbers enough to tell the full story of your impact? In this episode of the Common Good Data podcast, Drew Reynolds sits down with Cheralynn Corsack, founder of Local Insight Studio, to explore how mixed methods evaluation can produce deeper, more actionable insight, especially in rural communities.Evaluation conversations often center on numbers. Outputs. Outcomes. KPIs. But data alone rarely captures the nuance of lived experience. Cheralynn explains how pairing quantitative data with qualitative insight, including interviews, focus groups, and participatory analysis, reveals dimensions of impact that surveys alone cannot surface.The conversation explores:• What mixed methods evaluation actually means in practice• Why participatory approaches are especially powerful in rural communities• How qualitative insight can reshape and deepen quantitative findings• The challenges of data access and representation in rural contexts• Moving from deficit based narratives to asset based framing• Translating evaluation findings into language communities can understand and useCheralynn also discusses the importance of relationship building, trust, and co-creation in evaluation work, and why sharing findings back to communities is not optional but essential.If you work in nonprofits, philanthropy, or community initiatives and want your evaluation work to be rigorous, human centered, and useful, this episode offers practical insight you can apply immediately.Learn more about Cheralynn and Local Insight Studio at localinsightstudio.comExplore Common Good Data's free course, Break the Starvation Cycle, at commongooddata.com/coursesSubscribe for more conversations on evaluation, strategy, and data for social impact.

Learn more about Refrigeration Mentor Customized Technical Training Programs at www.refrigerationmentor.com/courses Join the Refrigeration Mentor Hub here This conversation was from our latest Refrigeration Mentor Community Meetup, talking about refrigeration controls and electrical systems with Andrew Freeburg and Erik Holland. We cover control fundamentals such as transformers, multiplex board setup, communication basics, polarity, baud rate, cable practices, and fail-safe settings for loads. We also discuss how to build confidence through competence - studying, repetition, applying skills on real systems, asking questions, using community support, setting goals, and learning by teaching. Interested in joining the next Refrigeration Mentor Community Meetup? Click here. In this episode, we discuss: (00:30) Confidence and Competence (06:02) Learning How to Learn (09:58) Setting Goals and Support Groups (15:42) Dunning Kruger Effect (21:58) Electrical Basics and Safety (22:21) Center Tap Transformers (24:30) Multiplex Boards and Dip Switches (25:59) Binary Addressing Switches (26:37) Power and Comms Terminals (27:11) Comms Voltage and LEDs (29:40) Wiring Noise and Shielding (30:47) Fail Safe Dip Switches (33:46) Analog Inputs and Outputs (34:54) Software vs Hardware Logic (39:06) Panel Safety Basics (43:30) Meter Testing and Ratings (47:47) Electrical Safety Mindset  Helpful Links & Resources: Episode 371. A 6-Step Process for Faster Electrical Troubleshooting Episode 215. Understanding Refrigeration System Controls with Larry Herman of Redline Control Design

The perfect AI storm happened, and no one has noticed yet.

Nothing speaks more to a changing world than our own mind and body. We are designed for change; we change every second of every day—naturally. If we choose to resist change, we fight nature and interfere with the innate flow of life. This is why John avoids the term “anti-aging.” When we fight aging, we age faster, grow tired, and give power to the very things we do not want. In this episode, John offers tips on how to relax into the "flow" of life to improve your quality of life—physically, mentally, emotionally, and spiritually. Now in his mid-60s, John shares how scientific measures recently placed his “internal age” at just 33 years.

Nothing speaks more to a changing world than our own mind and body. We are designed for change; we change every second of every day—naturally. If we choose to resist change, we fight nature and interfere with the innate flow of life. This is why John avoids the term “anti-aging.” When we fight aging, we age faster, grow tired, and give power to the very things we do not want. In this episode, John offers tips on how to relax into the "flow" of life to improve your quality of life—physically, mentally, emotionally, and spiritually. Now in his mid-60s, John shares how scientific measures recently placed his “internal age” at just 33 years.

How did prompt engineering die so quickly? ☠️And what the heck does context engineering even mean? One of the trickiest things about LLMs is they're changing daily, yet they're the engines that drive business results. But if the engine is constantly changing, then you also have to change how you drive and the roads you take. That's why we're tackling context engineering in this installment of our Start Here Series, the essential beginners guide to understanding AI basics and growing your skills. Context Engineering: How to Get Expert-Level Outputs From AI Chatbots -- An Everyday AI Chat with Jordan WilsonNewsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageJoin the discussion on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: info@youreverydayai.comConnect with Jordan on LinkedInTopics Covered in This Episode:Evolution from Prompt to Context EngineeringWhy Prompt Engineering Is Now ObsoleteDefining Context Engineering in AI ChatbotsSix-Part Framework for Context EngineeringFour Layer System for Structuring AI ContextBuilding Reusable Context Vaults and SkillsConnecting Business Data to AI ModelsTechniques to Achieve Expert-Level AI OutputsImportance of Context Windows in Large Language ModelsContext Engineering Best Practices and ScalabilityTimestamps:00:00 "Access AI Community & Tools"03:08 "Mastering Context in AI"07:23 "Smart Models Require Less Precision"12:01 "Context Engineering Beats Prompt Engineering"15:49 "AI Context: Six Key Blocks"16:47 "Building Context for Better Results"19:53 "AI: Training, Not Easy Button"25:17 "Chain of Thought Prompting Decline"29:11 "Show, Don't Tell Techniques"32:13 "Context, Reuse, and Scalable Systems"33:19 "AI Chatbots: Memory and Skills"Keywords: context engineering, AI chatbots, expert level outputs, prompt engineering, large language models, business context, AI models, custom instructions, data access, context window, prime prompt polish, reusable context vaults, context vaults, skills file, memory enabled models, ChatGPT, Claude, Google Gemini, Microsoft Copilot, connectors, apps, searchable index, business data, personalized AI, context clues, reference material, examples, procedures, evaluation rubric, chain of thought prompting, generative AI, nondeterministic behavior, show don't tell technique, few shot examples, rubric first technique, grading criteria, output quality, scalable AI systems,Send Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info) Ready for ROI on GenAI? Go to youreverydayai.com/partner 

From Palantir and Two Sigma to building Goodfire into the poster-child for actionable mechanistic interpretability, Mark Bissell (Member of Technical Staff) and Myra Deng (Head of Product) are trying to turn “peeking inside the model” into a repeatable production workflow by shipping APIs, landing real enterprise deployments, and now scaling the bet with a recent $150M Series B funding round at a $1.25B valuation.In this episode, we go far beyond the usual “SAEs are cool” take. We talk about Goodfire's core bet: that the AI lifecycle is still fundamentally broken because the only reliable control we have is data and we post-train, RLHF, and fine-tune by “slurping supervision through a straw,” hoping the model picks up the right behaviors while quietly absorbing the wrong ones. Goodfire's answer is to build a bi-directional interface between humans and models: read what's happening inside, edit it surgically, and eventually use interpretability during training so customization isn't just brute-force guesswork.Mark and Myra walk through what that looks like when you stop treating interpretability like a lab demo and start treating it like infrastructure: lightweight probes that add near-zero latency, token-level safety filters that can run at inference time, and interpretability workflows that survive messy constraints (multilingual inputs, synthetic→real transfer, regulated domains, no access to sensitive data). We also get a live window into what “frontier-scale interp” means operationally (i.e. steering a trillion-parameter model in real time by targeting internal features) plus why the same tooling generalizes cleanly from language models to genomics, medical imaging, and “pixel-space” world models.We discuss:* Myra + Mark's path: Palantir (health systems, forward-deployed engineering) → Goodfire early team; Two Sigma → Head of Product, translating frontier interpretability research into a platform and real-world deployments* What “interpretability” actually means in practice: not just post-hoc poking, but a broader “science of deep learning” approach across the full AI lifecycle (data curation → post-training → internal representations → model design)* Why post-training is the first big wedge: “surgical edits” for unintended behaviors likereward hacking, sycophancy, noise learned during customization plus the dream of targeted unlearning and bias removal without wrecking capabilities* SAEs vs probes in the real world: why SAE feature spaces sometimes underperform classifiers trained on raw activations for downstream detection tasks (hallucination, harmful intent, PII), and what that implies about “clean concept spaces”* Rakuten in production: deploying interpretability-based token-level PII detection at inference time to prevent routing private data to downstream providers plus the gnarly constraints: no training on real customer PII, synthetic→real transfer, English + Japanese, and tokenization quirks* Why interp can be operationally cheaper than LLM-judge guardrails: probes are lightweight, low-latency, and don't require hosting a second large model in the loop* Real-time steering at frontier scale: a demo of steering Kimi K2 (~1T params) live and finding features via SAE pipelines, auto-labeling via LLMs, and toggling a “Gen-Z slang” feature across multiple layers without breaking tool use* Hallucinations as an internal signal: the case that models have latent uncertainty / “user-pleasing” circuitry you can detect and potentially mitigate more directly than black-box methods* Steering vs prompting: the emerging view that activation steering and in-context learning are more closely connected than people think, including work mapping between the two (even for jailbreak-style behaviors)* Interpretability for science: using the same tooling across domains (genomics, medical imaging, materials) to debug spurious correlations and extract new knowledge up to and including early biomarker discovery work with major partners* World models + “pixel-space” interpretability: why vision/video models make concepts easier to see, how that accelerates the feedback loop, and why robotics/world-model partners are especially interesting design partners* The north star: moving from “data in, weights out” to intentional model design where experts can impart goals and constraints directly, not just via reward signals and brute-force post-training—Goodfire AI* Website: https://goodfire.ai* LinkedIn: https://www.linkedin.com/company/goodfire-ai/* X: https://x.com/GoodfireAIMyra Deng* Website: https://myradeng.com/* LinkedIn: https://www.linkedin.com/in/myra-deng/* X: https://x.com/myra_dengMark Bissell* LinkedIn: https://www.linkedin.com/in/mark-bissell/* X: https://x.com/MarkMBissellFull Video EpisodeTimestamps00:00:00 Introduction00:00:05 Introduction to the Latent Space Podcast and Guests from Goodfire00:00:29 What is Goodfire? Mission and Focus on Interpretability00:01:01 Goodfire's Practical Approach to Interpretability00:01:37 Goodfire's Series B Fundraise Announcement00:02:04 Backgrounds of Mark and Myra from Goodfire00:02:51 Team Structure and Roles at Goodfire00:05:13 What is Interpretability? Definitions and Techniques00:05:30 Understanding Errors00:07:29 Post-training vs. Pre-training Interpretability Applications00:08:51 Using Interpretability to Remove Unwanted Behaviors00:10:09 Grokking, Double Descent, and Generalization in Models00:10:15 404 Not Found Explained00:12:06 Subliminal Learning and Hidden Biases in Models00:14:07 How Goodfire Chooses Research Directions and Projects00:15:00 Troubleshooting Errors00:16:04 Limitations of SAEs and Probes in Interpretability00:18:14 Rakuten Case Study: Production Deployment of Interpretability00:20:45 Conclusion00:21:12 Efficiency Benefits of Interpretability Techniques00:21:26 Live Demo: Real-Time Steering in a Trillion Parameter Model00:25:15 How Steering Features are Identified and Labeled00:26:51 Detecting and Mitigating Hallucinations Using Interpretability00:31:20 Equivalence of Activation Steering and Prompting00:34:06 Comparing Steering with Fine-Tuning and LoRA Techniques00:36:04 Model Design and the Future of Intentional AI Development00:38:09 Getting Started in Mechinterp: Resources, Programs, and Open Problems00:40:51 Industry Applications and the Rise of Mechinterp in Practice00:41:39 Interpretability for Code Models and Real-World Usage00:43:07 Making Steering Useful for More Than Stylistic Edits00:46:17 Applying Interpretability to Healthcare and Scientific Discovery00:49:15 Why Interpretability is Crucial in High-Stakes Domains like Healthcare00:52:03 Call for Design Partners Across Domains00:54:18 Interest in World Models and Visual Interpretability00:57:22 Sci-Fi Inspiration: Ted Chiang and Interpretability01:00:14 Interpretability, Safety, and Alignment Perspectives01:04:27 Weak-to-Strong Generalization and Future Alignment Challenges01:05:38 Final Thoughts and Hiring/Collaboration Opportunities at GoodfireTranscriptShawn Wang [00:00:05]: So welcome to the Latent Space pod. We're back in the studio with our special MechInterp co-host, Vibhu. Welcome. Mochi, Mochi's special co-host. And Mochi, the mechanistic interpretability doggo. We have with us Mark and Myra from Goodfire. Welcome. Thanks for having us on. Maybe we can sort of introduce Goodfire and then introduce you guys. How do you introduce Goodfire today?Myra Deng [00:00:29]: Yeah, it's a great question. So Goodfire, we like to say, is an AI research lab that focuses on using interpretability to understand, learn from, and design AI models. And we really believe that interpretability will unlock the new generation, next frontier of safe and powerful AI models. That's our description right now, and I'm excited to dive more into the work we're doing to make that happen.Shawn Wang [00:00:55]: Yeah. And there's always like the official description. Is there an understatement? Is there an unofficial one that sort of resonates more with a different audience?Mark Bissell [00:01:01]: Well, being an AI research lab that's focused on interpretability, there's obviously a lot of people have a lot that they think about when they think of interpretability. And I think we have a pretty broad definition of what that means and the types of places that can be applied. And in particular, applying it in production scenarios, in high stakes industries, and really taking it sort of from the research world into the real world. Which, you know. It's a new field, so that hasn't been done all that much. And we're excited about actually seeing that sort of put into practice.Shawn Wang [00:01:37]: Yeah, I would say it wasn't too long ago that Anthopic was like still putting out like toy models or superposition and that kind of stuff. And I wouldn't have pegged it to be this far along. When you and I talked at NeurIPS, you were talking a little bit about your production use cases and your customers. And then not to bury the lead, today we're also announcing the fundraise, your Series B. $150 million. $150 million at a 1.25B valuation. Congrats, Unicorn.Mark Bissell [00:02:02]: Thank you. Yeah, no, things move fast.Shawn Wang [00:02:04]: We were talking to you in December and already some big updates since then. Let's dive, I guess, into a bit of your backgrounds as well. Mark, you were at Palantir working on health stuff, which is really interesting because the Goodfire has some interesting like health use cases. I don't know how related they are in practice.Mark Bissell [00:02:22]: Yeah, not super related, but I don't know. It was helpful context to know what it's like. Just to work. Just to work with health systems and generally in that domain. Yeah.Shawn Wang [00:02:32]: And Mara, you were at Two Sigma, which actually I was also at Two Sigma back in the day. Wow, nice.Myra Deng [00:02:37]: Did we overlap at all?Shawn Wang [00:02:38]: No, this is when I was briefly a software engineer before I became a sort of developer relations person. And now you're head of product. What are your sort of respective roles, just to introduce people to like what all gets done in Goodfire?Mark Bissell [00:02:51]: Yeah, prior to Goodfire, I was at Palantir for about three years as a forward deployed engineer, now a hot term. Wasn't always that way. And as a technical lead on the health care team and at Goodfire, I'm a member of the technical staff. And honestly, that I think is about as specific as like as as I could describe myself because I've worked on a range of things. And, you know, it's it's a fun time to be at a team that's still reasonably small. I think when I joined one of the first like ten employees, now we're above 40, but still, it looks like there's always a mix of research and engineering and product and all of the above. That needs to get done. And I think everyone across the team is, you know, pretty, pretty switch hitter in the roles they do. So I think you've seen some of the stuff that I worked on related to image models, which was sort of like a research demo. More recently, I've been working on our scientific discovery team with some of our life sciences partners, but then also building out our core platform for more of like flexing some of the kind of MLE and developer skills as well.Shawn Wang [00:03:53]: Very generalist. And you also had like a very like a founding engineer type role.Myra Deng [00:03:58]: Yeah, yeah.Shawn Wang [00:03:59]: So I also started as I still am a member of technical staff, did a wide range of things from the very beginning, including like finding our office space and all of this, which is we both we both visited when you had that open house thing. It was really nice.Myra Deng [00:04:13]: Thank you. Thank you. Yeah. Plug to come visit our office.Shawn Wang [00:04:15]: It looked like it was like 200 people. It has room for 200 people. But you guys are like 10.Myra Deng [00:04:22]: For a while, it was very empty. But yeah, like like Mark, I spend. A lot of my time as as head of product, I think product is a bit of a weird role these days, but a lot of it is thinking about how do we take our frontier research and really apply it to the most important real world problems and how does that then translate into a platform that's repeatable or a product and working across, you know, the engineering and research teams to make that happen and also communicating to the world? Like, what is interpretability? What is it used for? What is it good for? Why is it so important? All of these things are part of my day-to-day as well.Shawn Wang [00:05:01]: I love like what is things because that's a very crisp like starting point for people like coming to a field. They all do a fun thing. Vibhu, why don't you want to try tackling what is interpretability and then they can correct us.Vibhu Sapra [00:05:13]: Okay, great. So I think like one, just to kick off, it's a very interesting role to be head of product, right? Because you guys, at least as a lab, you're more of an applied interp lab, right? Which is pretty different than just normal interp, like a lot of background research. But yeah. You guys actually ship an API to try these things. You have Ember, you have products around it, which not many do. Okay. What is interp? So basically you're trying to have an understanding of what's going on in model, like in the model, in the internal. So different approaches to do that. You can do probing, SAEs, transcoders, all this stuff. But basically you have an, you have a hypothesis. You have something that you want to learn about what's happening in a model internals. And then you're trying to solve that from there. You can do stuff like you can, you know, you can do activation mapping. You can try to do steering. There's a lot of stuff that you can do, but the key question is, you know, from input to output, we want to have a better understanding of what's happening and, you know, how can we, how can we adjust what's happening on the model internals? How'd I do?Mark Bissell [00:06:12]: That was really good. I think that was great. I think it's also a, it's kind of a minefield of a, if you ask 50 people who quote unquote work in interp, like what is interpretability, you'll probably get 50 different answers. And. Yeah. To some extent also like where, where good fire sits in the space. I think that we're an AI research company above all else. And interpretability is a, is a set of methods that we think are really useful and worth kind of specializing in, in order to accomplish the goals we want to accomplish. But I think we also sort of see some of the goals as even more broader as, as almost like the science of deep learning and just taking a not black box approach to kind of any part of the like AI development life cycle, whether that. That means using interp for like data curation while you're training your model or for understanding what happened during post-training or for the, you know, understanding activations and sort of internal representations, what is in there semantically. And then a lot of sort of exciting updates that were, you know, are sort of also part of the, the fundraise around bringing interpretability to training, which I don't think has been done all that much before. A lot of this stuff is sort of post-talk poking at models as opposed to. To actually using this to intentionally design them.Shawn Wang [00:07:29]: Is this post-training or pre-training or is that not a useful.Myra Deng [00:07:33]: Currently focused on post-training, but there's no reason the techniques wouldn't also work in pre-training.Shawn Wang [00:07:38]: Yeah. It seems like it would be more active, applicable post-training because basically I'm thinking like rollouts or like, you know, having different variations of a model that you can tweak with the, with your steering. Yeah.Myra Deng [00:07:50]: And I think in a lot of the news that you've seen in, in, on like Twitter or whatever, you've seen a lot of unintended. Side effects come out of post-training processes, you know, overly sycophantic models or models that exhibit strange reward hacking behavior. I think these are like extreme examples. There's also, you know, very, uh, mundane, more mundane, like enterprise use cases where, you know, they try to customize or post-train a model to do something and it learns some noise or it doesn't appropriately learn the target task. And a big question that we've always had is like, how do you use your understanding of what the model knows and what it's doing to actually guide the learning process?Shawn Wang [00:08:26]: Yeah, I mean, uh, you know, just to anchor this for people, uh, one of the biggest controversies of last year was 4.0 GlazeGate. I've never heard of GlazeGate. I didn't know that was what it was called. The other one, they called it that on the blog post and I was like, well, how did OpenAI call it? Like officially use that term. And I'm like, that's funny, but like, yeah, I guess it's the pitch that if they had worked a good fire, they wouldn't have avoided it. Like, you know what I'm saying?Myra Deng [00:08:51]: I think so. Yeah. Yeah.Mark Bissell [00:08:53]: I think that's certainly one of the use cases. I think. Yeah. Yeah. I think the reason why post-training is a place where this makes a lot of sense is a lot of what we're talking about is surgical edits. You know, you want to be able to have expert feedback, very surgically change how your model is doing, whether that is, you know, removing a certain behavior that it has. So, you know, one of the things that we've been looking at or is, is another like common area where you would want to make a somewhat surgical edit is some of the models that have say political bias. Like you look at Quen or, um, R1 and they have sort of like this CCP bias.Shawn Wang [00:09:27]: Is there a CCP vector?Mark Bissell [00:09:29]: Well, there's, there are certainly internal, yeah. Parts of the representation space where you can sort of see where that lives. Yeah. Um, and you want to kind of, you know, extract that piece out.Shawn Wang [00:09:40]: Well, I always say, you know, whenever you find a vector, a fun exercise is just like, make it very negative to see what the opposite of CCP is.Mark Bissell [00:09:47]: The super America, bald eagles flying everywhere. But yeah. So in general, like lots of post-training tasks where you'd want to be able to, to do that. Whether it's unlearning a certain behavior or, you know, some of the other kind of cases where this comes up is, are you familiar with like the, the grokking behavior? I mean, I know the machine learning term of grokking.Shawn Wang [00:10:09]: Yeah.Mark Bissell [00:10:09]: Sort of this like double descent idea of, of having a model that is able to learn a generalizing, a generalizing solution, as opposed to even if memorization of some task would suffice, you want it to learn the more general way of doing a thing. And so, you know, another. A way that you can think about having surgical access to a model's internals would be learn from this data, but learn in the right way. If there are many possible, you know, ways to, to do that. Can make interp solve the double descent problem?Shawn Wang [00:10:41]: Depends, I guess, on how you. Okay. So I, I, I viewed that double descent as a problem because then you're like, well, if the loss curves level out, then you're done, but maybe you're not done. Right. Right. But like, if you actually can interpret what is a generalizing or what you're doing. What is, what is still changing, even though the loss is not changing, then maybe you, you can actually not view it as a double descent problem. And actually you're just sort of translating the space in which you view loss and like, and then you have a smooth curve. Yeah.Mark Bissell [00:11:11]: I think that's certainly like the domain of, of problems that we're, that we're looking to get.Shawn Wang [00:11:15]: Yeah. To me, like double descent is like the biggest thing to like ML research where like, if you believe in scaling, then you don't need, you need to know where to scale. And. But if you believe in double descent, then you don't, you don't believe in anything where like anything levels off, like.Vibhu Sapra [00:11:30]: I mean, also tendentially there's like, okay, when you talk about the China vector, right. There's the subliminal learning work. It was from the anthropic fellows program where basically you can have hidden biases in a model. And as you distill down or, you know, as you train on distilled data, those biases always show up, even if like you explicitly try to not train on them. So, you know, it's just like another use case of. Okay. If we can interpret what's happening in post-training, you know, can we clear some of this? Can we even determine what's there? Because yeah, it's just like some worrying research that's out there that shows, you know, we really don't know what's going on.Mark Bissell [00:12:06]: That is. Yeah. I think that's the biggest sentiment that we're sort of hoping to tackle. Nobody knows what's going on. Right. Like subliminal learning is just an insane concept when you think about it. Right. Train a model on not even the logits, literally the output text of a bunch of random numbers. And now your model loves owls. And you see behaviors like that, that are just, they defy, they defy intuition. And, and there are mathematical explanations that you can get into, but. I mean.Shawn Wang [00:12:34]: It feels so early days. Objectively, there are a sequence of numbers that are more owl-like than others. There, there should be.Mark Bissell [00:12:40]: According to, according to certain models. Right. It's interesting. I think it only applies to models that were initialized from the same starting Z. Usually, yes.Shawn Wang [00:12:49]: But I mean, I think that's a, that's a cheat code because there's not enough compute. But like if you believe in like platonic representation, like probably it will transfer across different models as well. Oh, you think so?Mark Bissell [00:13:00]: I think of it more as a statistical artifact of models initialized from the same seed sort of. There's something that is like path dependent from that seed that might cause certain overlaps in the latent space and then sort of doing this distillation. Yeah. Like it pushes it towards having certain other tendencies.Vibhu Sapra [00:13:24]: Got it. I think there's like a bunch of these open-ended questions, right? Like you can't train in new stuff during the RL phase, right? RL only reorganizes weights and you can only do stuff that's somewhat there in your base model. You're not learning new stuff. You're just reordering chains and stuff. But okay. My broader question is when you guys work at an interp lab, how do you decide what to work on and what's kind of the thought process? Right. Because we can ramble for hours. Okay. I want to know this. I want to know that. But like, how do you concretely like, you know, what's the workflow? Okay. There's like approaches towards solving a problem, right? I can try prompting. I can look at chain of thought. I can train probes, SAEs. But how do you determine, you know, like, okay, is this going anywhere? Like, do we have set stuff? Just, you know, if you can help me with all that. Yeah.Myra Deng [00:14:07]: It's a really good question. I feel like we've always at the very beginning of the company thought about like, let's go and try to learn what isn't working in machine learning today. Whether that's talking to customers or talking to researchers at other labs, trying to understand both where the frontier is going and where things are really not falling apart today. And then developing a perspective on how we can push the frontier using interpretability methods. And so, you know, even our chief scientist, Tom, spends a lot of time talking to customers and trying to understand what real world problems are and then taking that back and trying to apply the current state of the art to those problems and then seeing where they fall down basically. And then using those failures or those shortcomings to understand what hills to climb when it comes to interpretability research. So like on the fundamental side, for instance, when we have done some work applying SAEs and probes, we've encountered, you know, some shortcomings in SAEs that we found a little bit surprising. And so have gone back to the drawing board and done work on that. And then, you know, we've done some work on better foundational interpreter models. And a lot of our team's research is focused on what is the next evolution beyond SAEs, for instance. And then when it comes to like control and design of models, you know, we tried steering with our first API and realized that it still fell short of black box techniques like prompting or fine tuning. And so went back to the drawing board and we're like, how do we make that not the case and how do we improve it beyond that? And one of our researchers, Ekdeep, who just joined is actually Ekdeep and Atticus are like steering experts and have spent a lot of time trying to figure out like, what is the research that enables us to actually do this in a much more powerful, robust way? So yeah, the answer is like, look at real world problems, try to translate that into a research agenda and then like hill climb on both of those at the same time.Shawn Wang [00:16:04]: Yeah. Mark has the steering CLI demo queued up, which we're going to go into in a sec. But I always want to double click on when you drop hints, like we found some problems with SAEs. Okay. What are they? You know, and then we can go into the demo. Yeah.Myra Deng [00:16:19]: I mean, I'm curious if you have more thoughts here as well, because you've done it in the healthcare domain. But I think like, for instance, when we do things like trying to detect behaviors within models that are harmful or like behaviors that a user might not want to have in their model. So hallucinations, for instance, harmful intent, PII, all of these things. We first tried using SAE probes for a lot of these tasks. So taking the feature activation space from SAEs and then training classifiers on top of that, and then seeing how well we can detect the properties that we might want to detect in model behavior. And we've seen in many cases that probes just trained on raw activations seem to perform better than SAE probes, which is a bit surprising if you think that SAEs are actually also capturing the concepts that you would want to capture cleanly and more surgically. And so that is an interesting observation. I don't think that is like, I'm not down on SAEs at all. I think there are many, many things they're useful for, but we have definitely run into cases where I think the concept space described by SAEs is not as clean and accurate as we would expect it to be for actual like real world downstream performance metrics.Mark Bissell [00:17:34]: Fair enough. Yeah. It's the blessing and the curse of unsupervised methods where you get to peek into the AI's mind. But sometimes you wish that you saw other things when you walked inside there. Although in the PII instance, I think weren't an SAE based approach actually did prove to be the most generalizable?Myra Deng [00:17:53]: It did work well in the case that we published with Rakuten. And I think a lot of the reasons it worked well was because we had a noisier data set. And so actually the blessing of unsupervised learning is that we actually got to get more meaningful, generalizable signal from SAEs when the data was noisy. But in other cases where we've had like good data sets, it hasn't been the case.Shawn Wang [00:18:14]: And just because you named Rakuten and I don't know if we'll get it another chance, like what is the overall, like what is Rakuten's usage or production usage? Yeah.Myra Deng [00:18:25]: So they are using us to essentially guardrail and inference time monitor their language model usage and their agent usage to detect things like PII so that they don't route private user information.Myra Deng [00:18:41]: And so that's, you know, going through all of their user queries every day. And that's something that we deployed with them a few months ago. And now we are actually exploring very early partnerships, not just with Rakuten, but with other people around how we can help with potentially training and customization use cases as well. Yeah.Shawn Wang [00:19:03]: And for those who don't know, like it's Rakuten is like, I think number one or number two e-commerce store in Japan. Yes. Yeah.Mark Bissell [00:19:10]: And I think that use case actually highlights a lot of like what it looks like to deploy things in practice that you don't always think about when you're doing sort of research tasks. So when you think about some of the stuff that came up there that's more complex than your idealized version of a problem, they were encountering things like synthetic to real transfer of methods. So they couldn't train probes, classifiers, things like that on actual customer data of PII. So what they had to do is use synthetic data sets. And then hope that that transfer is out of domain to real data sets. And so we can evaluate performance on the real data sets, but not train on customer PII. So that right off the bat is like a big challenge. You have multilingual requirements. So this needed to work for both English and Japanese text. Japanese text has all sorts of quirks, including tokenization behaviors that caused lots of bugs that caused us to be pulling our hair out. And then also a lot of tasks you'll see. You might make simplifying assumptions if you're sort of treating it as like the easiest version of the problem to just sort of get like general results where maybe you say you're classifying a sentence to say, does this contain PII? But the need that Rakuten had was token level classification so that you could precisely scrub out the PII. So as we learned more about the problem, you're sort of speaking about what that looks like in practice. Yeah. A lot of assumptions end up breaking. And that was just one instance where you. A problem that seems simple right off the bat ends up being more complex as you keep diving into it.Vibhu Sapra [00:20:41]: Excellent. One of the things that's also interesting with Interp is a lot of these methods are very efficient, right? So where you're just looking at a model's internals itself compared to a separate like guardrail, LLM as a judge, a separate model. One, you have to host it. Two, there's like a whole latency. So if you use like a big model, you have a second call. Some of the work around like self detection of hallucination, it's also deployed for efficiency, right? So if you have someone like Rakuten doing it in production live, you know, that's just another thing people should consider.Mark Bissell [00:21:12]: Yeah. And something like a probe is super lightweight. Yeah. It's no extra latency really. Excellent.Shawn Wang [00:21:17]: You have the steering demos lined up. So we were just kind of see what you got. I don't, I don't actually know if this is like the latest, latest or like alpha thing.Mark Bissell [00:21:26]: No, this is a pretty hacky demo from from a presentation that someone else on the team recently gave. So this will give a sense for, for technology. So you can see the steering and action. Honestly, I think the biggest thing that this highlights is that as we've been growing as a company and taking on kind of more and more ambitious versions of interpretability related problems, a lot of that comes to scaling up in various different forms. And so here you're going to see steering on a 1 trillion parameter model. This is Kimi K2. And so it's sort of fun that in addition to the research challenges, there are engineering challenges that we're now tackling. Cause for any of this to be sort of useful in production, you need to be thinking about what it looks like when you're using these methods on frontier models as opposed to sort of like toy kind of model organisms. So yeah, this was thrown together hastily, pretty fragile behind the scenes, but I think it's quite a fun demo. So screen sharing is on. So I've got two terminal sessions pulled up here. On the left is a forked version that we have of the Kimi CLI that we've got running to point at our custom hosted Kimi model. And then on the right is a set up that will allow us to steer on certain concepts. So I should be able to chat with Kimi over here. Tell it hello. This is running locally. So the CLI is running locally, but the Kimi server is running back to the office. Well, hopefully should be, um, that's too much to run on that Mac. Yeah. I think it's, uh, it takes a full, like each 100 node. I think it's like, you can. You can run it on eight GPUs, eight 100. So, so yeah, Kimi's running. We can ask it a prompt. It's got a forked version of our, uh, of the SG line code base that we've been working on. So I'm going to tell it, Hey, this SG line code base is slow. I think there's a bug. Can you try to figure it out? There's a big code base, so it'll, it'll spend some time doing this. And then on the right here, I'm going to initialize in real time. Some steering. Let's see here.Mark Bissell [00:23:33]: searching for any. Bugs. Feature ID 43205.Shawn Wang [00:23:38]: Yeah.Mark Bissell [00:23:38]: 20, 30, 40. So let me, uh, this is basically a feature that we found that inside Kimi seems to cause it to speak in Gen Z slang. And so on the left, it's still sort of thinking normally it might take, I don't know, 15 seconds for this to kick in, but then we're going to start hopefully seeing him do this code base is massive for real. So we're going to start. We're going to start seeing Kimi transition as the steering kicks in from normal Kimi to Gen Z Kimi and both in its chain of thought and its actual outputs.Mark Bissell [00:24:19]: And interestingly, you can see, you know, it's still able to call tools, uh, and stuff. It's um, it's purely sort of it's it's demeanor. And there are other features that we found for interesting things like concision. So that's more of a practical one. You can make it more concise. Um, the types of programs, uh, programming languages that uses, but yeah, as we're seeing it come in. Pretty good. Outputs.Shawn Wang [00:24:43]: Scheduler code is actually wild.Vibhu Sapra [00:24:46]: Yo, this code is actually insane, bro.Vibhu Sapra [00:24:53]: What's the process of training in SAE on this, or, you know, how do you label features? I know you guys put out a pretty cool blog post about, um, finding this like autonomous interp. Um, something. Something about how agents for interp is different than like coding agents. I don't know while this is spewing up, but how, how do we find feature 43, two Oh five. Yeah.Mark Bissell [00:25:15]: So in this case, um, we, our platform that we've been building out for a long time now supports all the sort of classic out of the box interp techniques that you might want to have like SAE training, probing things of that kind, I'd say the techniques for like vanilla SAEs are pretty well established now where. You take your model that you're interpreting, run a whole bunch of data through it, gather activations, and then yeah, pretty straightforward pipeline to train an SAE. There are a lot of different varieties. There's top KSAEs, batch top KSAEs, um, normal ReLU SAEs. And then once you have your sparse features to your point, assigning labels to them to actually understand that this is a gen Z feature, that's actually where a lot of the kind of magic happens. Yeah. And the most basic standard technique is look at all of your d input data set examples that cause this feature to fire most highly. And then you can usually pick out a pattern. So for this feature, If I've run a diverse enough data set through my model feature 43, two Oh five. Probably tends to fire on all the tokens that sounds like gen Z slang. You know, that's the, that's the time of year to be like, Oh, I'm in this, I'm in this Um, and, um, so, you know, you could have a human go through all 43,000 concepts andVibhu Sapra [00:26:34]: And I've got to ask the basic question, you know, can we get examples where it hallucinates, pass it through, see what feature activates for hallucinations? Can I just, you know, turn hallucination down?Myra Deng [00:26:51]: Oh, wow. You really predicted a project we're already working on right now, which is detecting hallucinations using interpretability techniques. And this is interesting because hallucinations is something that's very hard to detect. And it's like a kind of a hairy problem and something that black box methods really struggle with. Whereas like Gen Z, you could always train a simple classifier to detect that hallucinations is harder. But we've seen that models internally have some... Awareness of like uncertainty or some sort of like user pleasing behavior that leads to hallucinatory behavior. And so, yeah, we have a project that's trying to detect that accurately. And then also working on mitigating the hallucinatory behavior in the model itself as well.Shawn Wang [00:27:39]: Yeah, I would say most people are still at the level of like, oh, I would just turn temperature to zero and that turns off hallucination. And I'm like, well, that's a fundamental misunderstanding of how this works. Yeah.Mark Bissell [00:27:51]: Although, so part of what I like about that question is you, there are SAE based approaches that might like help you get at that. But oftentimes the beauty of SAEs and like we said, the curse is that they're unsupervised. So when you have a behavior that you deliberately would like to remove, and that's more of like a supervised task, often it is better to use something like probes and specifically target the thing that you're interested in reducing as opposed to sort of like hoping that when you fragment the latent space, one of the vectors that pops out.Vibhu Sapra [00:28:20]: And as much as we're training an autoencoder to be sparse, we're not like for sure certain that, you know, we will get something that just correlates to hallucination. You'll probably split that up into 20 other things and who knows what they'll be.Mark Bissell [00:28:36]: Of course. Right. Yeah. So there's no sort of problems with like feature splitting and feature absorption. And then there's the off target effects, right? Ideally, you would want to be very precise where if you reduce the hallucination feature, suddenly maybe your model can't write. Creatively anymore. And maybe you don't like that, but you want to still stop it from hallucinating facts and figures.Shawn Wang [00:28:55]: Good. So Vibhu has a paper to recommend there that we'll put in the show notes. But yeah, I mean, I guess just because your demo is done, any any other things that you want to highlight or any other interesting features you want to show?Mark Bissell [00:29:07]: I don't think so. Yeah. Like I said, this is a pretty small snippet. I think the main sort of point here that I think is exciting is that there's not a whole lot of inter being applied to models quite at this scale. You know, Anthropic certainly has some some. Research and yeah, other other teams as well. But it's it's nice to see these techniques, you know, being put into practice. I think not that long ago, the idea of real time steering of a trillion parameter model would have sounded.Shawn Wang [00:29:33]: Yeah. The fact that it's real time, like you started the thing and then you edited the steering vector.Vibhu Sapra [00:29:38]: I think it's it's an interesting one TBD of what the actual like production use case would be on that, like the real time editing. It's like that's the fun part of the demo, right? You can kind of see how this could be served behind an API, right? Like, yes, you're you only have so many knobs and you can just tweak it a bit more. And I don't know how it plays in. Like people haven't done that much with like, how does this work with or without prompting? Right. How does this work with fine tuning? Like, there's a whole hype of continual learning, right? So there's just so much to see. Like, is this another parameter? Like, is it like parameter? We just kind of leave it as a default. We don't use it. So I don't know. Maybe someone here wants to put out a guide on like how to use this with prompting when to do what?Mark Bissell [00:30:18]: Oh, well, I have a paper recommendation. I think you would love from Act Deep on our team, who is an amazing researcher, just can't say enough amazing things about Act Deep. But he actually has a paper that as well as some others from the team and elsewhere that go into the essentially equivalence of activation steering and in context learning and how those are from a he thinks of everything in a cognitive neuroscience Bayesian framework, but basically how you can precisely show how. Prompting in context, learning and steering exhibit similar behaviors and even like get quantitative about the like magnitude of steering you would need to do to induce a certain amount of behavior similar to certain prompting, even for things like jailbreaks and stuff. It's a really cool paper. Are you saying steering is less powerful than prompting? More like you can almost write a formula that tells you how to convert between the two of them.Myra Deng [00:31:20]: And so like formally equivalent actually in the in the limit. Right.Mark Bissell [00:31:24]: So like one case study of this is for jailbreaks there. I don't know. Have you seen the stuff where you can do like many shot jailbreaking? You like flood the context with examples of the behavior. And the topic put out that paper.Shawn Wang [00:31:38]: A lot of people were like, yeah, we've been doing this, guys.Mark Bissell [00:31:40]: Like, yeah, what's in this in context learning and activation steering equivalence paper is you can like predict the number. Number of examples that you will need to put in there in order to jailbreak the model. That's cool. By doing steering experiments and using this sort of like equivalence mapping. That's cool. That's really cool. It's very neat. Yeah.Shawn Wang [00:32:02]: I was going to say, like, you know, I can like back rationalize that this makes sense because, you know, what context is, is basically just, you know, it updates the KV cache kind of and like and then every next token inference is still like, you know, the sheer sum of everything all the way. It's plus all the context. It's up to date. And you could, I guess, theoretically steer that with you probably replace that with your steering. The only problem is steering typically is on one layer, maybe three layers like like you did. So it's like not exactly equivalent.Mark Bissell [00:32:33]: Right, right. There's sort of you need to get precise about, yeah, like how you sort of define steering and like what how you're modeling the setup. But yeah, I've got the paper pulled up here. Belief dynamics reveal the dual nature. Yeah. The title is Belief Dynamics Reveal the Dual Nature of Incompetence. And it's an exhibition of the practical context learning and activation steering. So Eric Bigelow, Dan Urgraft on the who are doing fellowships at Goodfire, Ekt Deep's the final author there.Myra Deng [00:32:59]: I think actually to your question of like, what is the production use case of steering? I think maybe if you just think like one level beyond steering as it is today. Like imagine if you could adapt your model to be, you know, an expert legal reasoner. Like in almost real time, like very quickly. efficiently using human feedback or using like your semantic understanding of what the model knows and where it knows that behavior. I think that while it's not clear what the product is at the end of the day, it's clearly very valuable. Thinking about like what's the next interface for model customization and adaptation is a really interesting problem for us. Like we have heard a lot of people actually interested in fine-tuning an RL for open weight models in production. And so people are using things like Tinker or kind of like open source libraries to do that, but it's still very difficult to get models fine-tuned and RL'd for exactly what you want them to do unless you're an expert at model training. And so that's like something we'reShawn Wang [00:34:06]: looking into. Yeah. I never thought so. Tinker from Thinking Machines famously uses rank one LoRa. Is that basically the same as steering? Like, you know, what's the comparison there?Mark Bissell [00:34:19]: Well, so in that case, you are still applying updates to the parameters, right?Shawn Wang [00:34:25]: Yeah. You're not touching a base model. You're touching an adapter. It's kind of, yeah.Mark Bissell [00:34:30]: Right. But I guess it still is like more in parameter space then. I guess it's maybe like, are you modifying the pipes or are you modifying the water flowing through the pipes to get what you're after? Yeah. Just maybe one way.Mark Bissell [00:34:44]: I like that analogy. That's my mental map of it at least, but it gets at this idea of model design and intentional design, which is something that we're, that we're very focused on. And just the fact that like, I hope that we look back at how we're currently training models and post-training models and just think what a primitive way of doing that right now. Like there's no intentionalityShawn Wang [00:35:06]: really in... It's just data, right? The only thing in control is what data we feed in.Mark Bissell [00:35:11]: So, so Dan from Goodfire likes to use this analogy of, you know, he has a couple of young kids and he talks about like, what if I could only teach my kids how to be good people by giving them cookies or like, you know, giving them a slap on the wrist if they do something wrong, like not telling them why it was wrong or like what they should have done differently or something like that. Just figure it out. Right. Exactly. So that's RL. Yeah. Right. And, and, you know, it's sample inefficient. There's, you know, what do they say? It's like slurping feedback. It's like, slurping supervision. Right. And so you'd like to get to the point where you can have experts giving feedback to their models that are, uh, internalized and, and, you know, steering is an inference time way of sort of getting that idea. But ideally you're moving to a world whereVibhu Sapra [00:36:04]: it is much more intentional design in perpetuity for these models. Okay. This is one of the questions we asked Emmanuel from Anthropic on the podcast a few months ago. Basically the question, was you're at a research lab that does model training, foundation models, and you're on an interp team. How does it tie back? Right? Like, does this, do ideas come from the pre-training team? Do they go back? Um, you know, so for those interested, you can, you can watch that. There wasn't too much of a connect there, but it's still something, you know, it's something they want toMark Bissell [00:36:33]: push for down the line. It can be useful for all of the above. Like there are certainly post-hocVibhu Sapra [00:36:39]: use cases where it doesn't need to touch that. I think the other thing a lot of people forget is this stuff isn't too computationally expensive, right? Like I would say, if you're interested in getting into research, MechInterp is one of the most approachable fields, right? A lot of this train an essay, train a probe, this stuff, like the budget for this one, there's already a lot done. There's a lot of open source work. You guys have done some too. Um, you know,Shawn Wang [00:37:04]: There's like notebooks from the Gemini team for Neil Nanda or like, this is how you do it. Just step through the notebook.Vibhu Sapra [00:37:09]: Even if you're like, not even technical with any of this, you can still make like progress. There, you can look at different activations, but, uh, if you do want to get into training, you know, training this stuff, correct me if I'm wrong is like in the thousands of dollars, not even like, it's not that high scale. And then same with like, you know, applying it, doing it for post-training or all this stuff is fairly cheap in scale of, okay. I want to get into like model training. I don't have compute for like, you know, pre-training stuff. So it's, it's a very nice field to get into. And also there's a lot of like open questions, right? Um, some of them have to go with, okay, I want a product. I want to solve this. Like there's also just a lot of open-ended stuff that people could work on. That's interesting. Right. I don't know if you guys have any calls for like, what's open questions, what's open work that you either open collaboration with, or like, you'd just like to see solved or just, you know, for people listening that want to get into McInturk because people always talk about it. What are, what are the things they should check out? Start, of course, you know, join you guys as well. I'm sure you're hiring.Myra Deng [00:38:09]: There's a paper, I think from, was it Lee, uh, Sharky? It's open problems and, uh, it's, it's a bit of interpretability, which I recommend everyone who's interested in the field. Read. I'm just like a really comprehensive overview of what are the things that experts in the field think are the most important problems to be solved. I also think to your point, it's been really, really inspiring to see, I think a lot of young people getting interested in interpretability, actually not just young people also like scientists to have been, you know, experts in physics for many years and in biology or things like this, um, transitioning into interp, because the barrier of, of what's now interp. So it's really cool to see a number to entry is, you know, in some ways low and there's a lot of information out there and ways to get started. There's this anecdote of like professors at universities saying that all of a sudden every incoming PhD student wants to study interpretability, which was not the case a few years ago. So it just goes to show how, I guess, like exciting the field is, how fast it's moving, how quick it is to get started and things like that.Mark Bissell [00:39:10]: And also just a very welcoming community. You know, there's an open source McInturk Slack channel. There are people are always posting questions and just folks in the space are always responsive if you ask things on various forums and stuff. But yeah, the open paper, open problems paper is a really good one.Myra Deng [00:39:28]: For other people who want to get started, I think, you know, MATS is a great program. What's the acronym for? Machine Learning and Alignment Theory Scholars? It's like the...Vibhu Sapra [00:39:40]: Normally summer internship style.Myra Deng [00:39:42]: Yeah, but they've been doing it year round now. And actually a lot of our full-time staff have come through that program or gone through that program. And it's great for anyone who is transitioning into interpretability. There's a couple other fellows programs. We do one as well as Anthropic. And so those are great places to get started if anyone is interested.Mark Bissell [00:40:03]: Also, I think been seen as a research field for a very long time. But I think engineering... I think engineers are sorely wanted for interpretability as well, especially at Goodfire, but elsewhere, as it does scale up.Shawn Wang [00:40:18]: I should mention that Lee actually works with you guys, right? And in the London office and I'm adding our first ever McInturk track at AI Europe because I see this industry applications now emerging. And I'm pretty excited to, you know, help push that along. Yeah, I was looking forward to that. It'll effectively be the first industry McInturk conference. Yeah. I'm so glad you added that. You know, it's still a little bit of a bet. It's not that widespread, but I can definitely see this is the time to really get into it. We want to be early on things.Mark Bissell [00:40:51]: For sure. And I think the field understands this, right? So at ICML, I think the title of the McInturk workshop this year was actionable interpretability. And there was a lot of discussion around bringing it to various domains. Everyone's adding pragmatic, actionable, whatever.Shawn Wang [00:41:10]: It's like, okay, well, we weren't actionable before, I guess. I don't know.Vibhu Sapra [00:41:13]: And I mean, like, just, you know, being in Europe, you see the Interp room. One, like old school conferences, like, I think they had a very tiny room till they got lucky and they got it doubled. But there's definitely a lot of interest, a lot of niche research. So you see a lot of research coming out of universities, students. We covered the paper last week. It's like two unknown authors, not many citations. But, you know, you can make a lot of meaningful work there. Yeah. Yeah. Yeah.Shawn Wang [00:41:39]: Yeah. I think people haven't really mentioned this yet. It's just Interp for code. I think it's like an abnormally important field. We haven't mentioned this yet. The conspiracy theory last two years ago was when the first SAE work came out of Anthropic was they would do like, oh, we just used SAEs to turn the bad code vector down and then turn up the good code. And I think like, isn't that the dream? Like, you know, like, but basically, I guess maybe, why is it funny? Like, it's... If it was realistic, it would not be funny. It would be like, no, actually, we should do this. But it's funny because we know there's like, we feel there's some limitations to what steering can do. And I think a lot of the public image of steering is like the Gen Z stuff. Like, oh, you can make it really love the Golden Gate Bridge, or you can make it speak like Gen Z. To like be a legal reasoner seems like a huge stretch. Yeah. And I don't know if that will get there this way. Yeah.Myra Deng [00:42:36]: I think, um, I will say we are announcing. Something very soon that I will not speak too much about. Um, but I think, yeah, this is like what we've run into again and again is like, we, we don't want to be in the world where steering is only useful for like stylistic things. That's definitely not, not what we're aiming for. But I think the types of interventions that you need to do to get to things like legal reasoning, um, are much more sophisticated and require breakthroughs in, in learning algorithms. And that's, um...Shawn Wang [00:43:07]: And is this an emergent property of scale as well?Myra Deng [00:43:10]: I think so. Yeah. I mean, I think scale definitely helps. I think scale allows you to learn a lot of information and, and reduce noise across, you know, large amounts of data. But I also think we think that there's ways to do things much more effectively, um, even, even at scale. So like actually learning exactly what you want from the data and not learning things that you do that you don't want exhibited in the data. So we're not like anti-scale, but we are also realizing that scale is not going to get us anywhere. It's not going to get us to the type of AI development that we want to be at in, in the future as these models get more powerful and get deployed in all these sorts of like mission critical contexts. Current life cycle of training and deploying and evaluations is, is to us like deeply broken and has opportunities to, to improve. So, um, more to come on that very, very soon.Mark Bissell [00:44:02]: And I think that that's a use basically, or maybe just like a proof point that these concepts do exist. Like if you can manipulate them in the precise best way, you can get the ideal combination of them that you desire. And steering is maybe the most coarse grained sort of peek at what that looks like. But I think it's evocative of what you could do if you had total surgical control over every concept, every parameter. Yeah, exactly.Myra Deng [00:44:30]: There were like bad code features. I've got it pulled up.Vibhu Sapra [00:44:33]: Yeah. Just coincidentally, as you guys are talking.Shawn Wang [00:44:35]: This is like, this is exactly.Vibhu Sapra [00:44:38]: There's like specifically a code error feature that activates and they show, you know, it's not, it's not typo detection. It's like, it's, it's typos in code. It's not typical typos. And, you know, you can, you can see it clearly activates where there's something wrong in code. And they have like malicious code, code error. They have a whole bunch of sub, you know, sub broken down little grain features. Yeah.Shawn Wang [00:45:02]: Yeah. So, so the, the rough intuition for me, the, why I talked about post-training was that, well, you just, you know, have a few different rollouts with all these things turned off and on and whatever. And then, you know, you can, that's, that's synthetic data you can kind of post-train on. Yeah.Vibhu Sapra [00:45:13]: And I think we make it sound easier than it is just saying, you know, they do the real hard work.Myra Deng [00:45:19]: I mean, you guys, you guys have the right idea. Exactly. Yeah. We replicated a lot of these features in, in our Lama models as well. I remember there was like.Vibhu Sapra [00:45:26]: And I think a lot of this stuff is open, right? Like, yeah, you guys opened yours. DeepMind has opened a lot of essays on Gemma. Even Anthropic has opened a lot of this. There's, there's a lot of resources that, you know, we can probably share of people that want to get involved.Shawn Wang [00:45:41]: Yeah. And special shout out to like Neuronpedia as well. Yes. Like, yeah, amazing piece of work to visualize those things.Myra Deng [00:45:49]: Yeah, exactly.Shawn Wang [00:45:50]: I guess I wanted to pivot a little bit on, onto the healthcare side, because I think that's a big use case for you guys. We haven't really talked about it yet. This is a bit of a crossover for me because we are, we are, we do have a separate science pod that we're starting up for AI, for AI for science, just because like, it's such a huge investment category and also I'm like less qualified to do it, but we actually have bio PhDs to cover that, which is great, but I need to just kind of recover, recap your work, maybe on the evil two stuff, but then, and then building forward.Mark Bissell [00:46:17]: Yeah, for sure. And maybe to frame up the conversation, I think another kind of interesting just lens on interpretability in general is a lot of the techniques that were described. are ways to solve the AI human interface problem. And it's sort of like bidirectional communication is the goal there. So what we've been talking about with intentional design of models and, you know, steering, but also more advanced techniques is having humans impart our desires and control into models and over models. And the reverse is also very interesting, especially as you get to superhuman models, whether that's narrow superintelligence, like these scientific models that work on genomics, data, medical imaging, things like that. But down the line, you know, superintelligence of other forms as well. What knowledge can the AIs teach us as sort of that, that the other direction in that? And so some of our life science work to date has been getting at exactly that question, which is, well, some of it does look like debugging these various life sciences models, understanding if they're actually performing well, on tasks, or if they're picking up on spurious correlations, for instance, genomics models, you would like to know whether they are sort of focusing on the biologically relevant things that you care about, or if it's using some simpler correlate, like the ancestry of the person that it's looking at. But then also in the instances where they are superhuman, and maybe they are understanding elements of the human genome that we don't have names for or specific, you know, yeah, discoveries that they've made that that we don't know about, that's, that's a big goal. And so we're already seeing that, right, we are partnered with organizations like Mayo Clinic, leading research health system in the United States, our Institute, as well as a startup called Prima Menta, which focuses on neurodegenerative disease. And in our partnership with them, we've used foundation models, they've been training and applied our interpretability techniques to find novel biomarkers for Alzheimer's disease. So I think this is just the tip of the iceberg. But it's, that's like a flavor of some of the things that we're working on.Shawn Wang [00:48:36]: Yeah, I think that's really fantastic. Obviously, we did the Chad Zuckerberg pod last year as well. And like, there's a plethora of these models coming out, because there's so much potential and research. And it's like, very interesting how it's basically the same as language models, but just with a different underlying data set. But it's like, it's the same exact techniques. Like, there's no change, basically.Mark Bissell [00:48:59]: Yeah. Well, and even in like other domains, right? Like, you know, robotics, I know, like a lot of the companies just use Gemma as like the like backbone, and then they like make it into a VLA that like takes these actions. It's, it's, it's transformers all the way down. So yeah.Vibhu Sapra [00:49:15]: Like we have Med Gemma now, right? Like this week, even there was Med Gemma 1.5. And they're training it on this stuff, like 3d scans, medical domain knowledge, and all that stuff, too. So there's a push from both sides. But I think the thing that, you know, one of the things about McInturpp is like, you're a little bit more cautious in some domains, right? So healthcare, mainly being one, like guardrails, understanding, you know, we're more risk adverse to something going wrong there. So even just from a basic understanding, like, if we're trusting these systems to make claims, we want to know why and what's going on.Myra Deng [00:49:51]: Yeah, I think there's totally a kind of like deployment bottleneck to actually using. foundation models for real patient usage or things like that. Like, say you're using a model for rare disease prediction, you probably want some explanation as to why your model predicted a certain outcome, and an interpretable explanation at that. So that's definitely a use case. But I also think like, being able to extract scientific information that no human knows to accelerate drug discovery and disease treatment and things like that actually is a really, really big unlock for science, like scientific discovery. And you've seen a lot of startups, like say that they're going to accelerate scientific discovery. And I feel like we actually are doing that through our interp techniques. And kind of like, almost by accident, like, I think we got reached out to very, very early on from these healthcare institutions. And none of us had healthcare.Shawn Wang [00:50:49]: How did they even hear of you? A podcast.Myra Deng [00:50:51]: Oh, okay. Yeah, podcast.Vibhu Sapra [00:50:53]: Okay, well, now's that time, you know.Myra Deng [00:50:55]: Everyone can call us.Shawn Wang [00:50:56]: Podcasts are the most important thing. Everyone should listen to podcasts.Myra Deng [00:50:59]: Yeah, they reached out. They were like, you know, we have these really smart models that we've trained, and we want to know what they're doing. And we were like, really early that time, like three months old, and it was a few of us. And we were like, oh, my God, we've never used these models. Let's figure it out. But it's also like, great proof that interp techniques scale pretty well across domains. We didn't really have to learn too much about.Shawn Wang [00:51:21]: Interp is a machine learning technique, machine learning skills everywhere, right? Yeah. And it's obviously, it's just like a general insight. Yeah. Probably to finance too, I think, which would be fun for our history. I don't know if you have anything to say there.Mark Bissell [00:51:34]: Yeah, well, just across the science. Like, we've also done work on material science. Yeah, it really runs the gamut.Vibhu Sapra [00:51:40]: Yeah. Awesome. And, you know, for those that should reach out, like, you're obviously experts in this, but like, is there a call out for people that you're looking to partner with, design partners, people to use your stuff outside of just, you know, the general developer that wants to. Plug and play steering stuff, like on the research side more so, like, are there ideal design partners, customers, stuff like that?Myra Deng [00:52:03]: Yeah, I can talk about maybe non-life sciences, and then I'm curious to hear from you on the life sciences side. But we're looking for design partners across many domains, language, anyone who's customizing language models or trying to push the frontier of code or reasoning models is really interesting to us. And then also interested in the frontier of modeling. There's a lot of models that work in, like, pixel space, as we call it. So if you're doing world models, video models, even robotics, where there's not a very clean natural language interface to interact with, I think we think that Interp can really help and are looking for a few partners in that space.Shawn Wang [00:52:43]: Just because you mentioned the keyword

Outcomes, not Outputs We wonder why Agile is dying. Should we be so surprised? You’re probably watching your wallet these days. Prices are up, the economy is slow and uncertain. Your employers are no different. In 2026, story points, velocity, lead time and throughput might matter to YOU, but they don’t matter to THEM. Not to say that those measures aren’t important. Heck, you can’t tune your development, testing and deployment operations if you’re not looking at them. But if you were called to the mat tomorrow, if you were asked to prove the ROI of what you do, those metrics don’t tell a compelling story. It’s about Outcomes, not Outputs. Coaches and SM’s are now being called to defend their worth. They’re not interested in how well-tuned the development machine is if that doesn’t translate to real business results. The measures that the business cares about have to do with their financial outcomes, not outputs that don’t face the customer. Are your efforts helping the company create shareholder value? Do they impact earnings-per-share? If your boss spends a few million on a cadre of Agile Coaches this year, is that a net-positive investment for the company? More cashflow? More customers? Fewer abandoned carts? New subscribers? I know. I sound super-corporate right now. And maybe you hate that. But if you’re looking to accelerate your career in 2026, you can’t ask the people who fund it to trust you, sight unseen. They’re taking a closer look at the books, and these questions are long overdue. How are you contributing to our results? The Dev team might be considered a necessary expense, but if your Agile Shirpa talent aren’t making the team more impactful than they would be on their own, why are you even here? Remember, its Outcomes, not Outputs. This is the future of Agile practice in large enterprise. You have to collaborate with the business, and drive results for them. If you want to learn how, you should check out my brand-new Business Outcomes Partner Playbook. Get the edge so you can get your career back where it belongs, and say good-bye to to the upheaval and uncertainty that’s ripping through our industry. Did you enjoy this episode? You might also like these: The 2026 18th State Of Agile Report Episode 224 – Circulate Value – The Agile Survival Skill Episode 235 – Agile Is A Doorway **LEARN HOW TO DELIVER UNDENIABLE ROI THAT SAVES YOUR JOB AND ACCELERATES YOUR FUTURE** Get the Business Outcomes Partner Playbook Now! https://learning.fusechamber.com/offers/AFGm3tSy/checkout **FORGE GENESIS IS HERE** All the skills you need to stop relying on job postings and start enjoying the freedom of an Agile career on YOUR terms. First cohort starts in Jan 2026 https://learning.fusechamber.com/forge-genesis **THE ALL NEW FORGE LIGHTNING** 12 Weeks to elite leadership! https://learning.fusechamber.com/forge-lightning **JOIN MY BETA COMMUNITY FOR AGILE ENTREPRENEURS AND INTRAPRENEURS** The latest wave in professional Agile careers. Get the support you need to Forge Your Freedom! Join for FREE here: https://learning.fusechamber.com/offers/Sa3udEgz **CHECK OUT ALL MY PRODUCTS AND SERVICES HERE:** https://learning.fusechamber.com **ELEVATE YOUR PROFESSIONAL STORYTELLING – Now Live!** The most coveted communications skill – now at your fingertips! https://learning.fusechamber.com/storytelling **JOIN THE FORGE*** New cohorts for Fall 2025!  Email for more information: contact@badassagile.com **BREAK FREE OF CORPORATE AGILE!!*** Download my FREE Guide and learn how to shift from roles and process and use your agile skills in new and exciting ways! https://learning.fusechamber.com/future-of-agile-signup We’re also on YouTube! Follow the podcast, enjoy some panel/guest commentary, and get some quick tips and guidance from me: https://www.youtube.com/c/BadassAgile ****** Follow The LinkedIn Page: https://www.linkedin.com/showcase/badass-agile ****** Our mission is to create an elite tribe of leaders who focus on who they need to become in order to lead and inspire, and to be the best agile podcast and resource for effective mindset and leadership game. Contact us (contact@badassagile.com) for elite-level performance and agile coaching, speaking engagements, team-level and executive mindset/agile training, and licensing options for modern, high-impact, bite-sized learning and educational content.

AI was supposed to help humans think better, decide better, and operate with more agency. Instead, many of us feel slower, less confident, and strangely replaceable.In this episode of Design of AI, we interviewed Ovetta Sampson about what quietly went wrong. Not in theory—in practice. We examine how frictionless tools displaced intention, how “freedom” became confused with unlimited capability, and how responsibility dissolved behind abstraction layers, vendors, and models no one fully controls.This is not an anti-AI conversation. It's a reckoning with what happens when adoption outruns judgment.Ovetta Sampson is a tech industry leader who has spent more than a decade leading engineers, designers, and researchers across some of the most influential organizations in technology, including Google, Microsoft, IDEO, and Capital One. She has designed and delivered machine learning, artificial intelligence, and enterprise software systems across multiple industries, and in 2023 was named one of Business Insider's Top 15 People in Enterprise Artificial Intelligence.Join her mailing list⁠ | Right AI | Free Mindful AI Playbook Why 2026 Will Force Teams to Rethink How Much AI They Actually NeedThe risks are no longer abstract. The tradeoffs are no longer subtle. Teams are already feeling the consequences: bloated tool stacks, degraded judgment, unclear accountability, and productivity that looks impressive but feels empty.The next advantage will not come from adding more AI. It will come from removing it deliberately.Organizations that adapt will narrow where AI is used—essential systems, bounded experiments, and clearly protected human decision points. The payoff won't just be cost savings. It will be the return of clarity, ownership, and trust. This is going to manifest first with individuals and small startups who were early adopters of AI. My prediction is that this year they'll start cutting the number of AI models they pay for because the era of experimentation is over and we're now entering a period where deliberate choices will matter more than how fast the model is. Read the full article on LinkedIn. Do You Really Need Frontier Models for Your Product to Work?For most teams, the honest answer is no.Open-source and on-device models already cover the majority of real business needs: internal tooling, retrieval, summarization, classification, workflow automation, and privacy-sensitive systems. The capability gap is routinely overstated—often by those selling access.What open models offer instead is control: over data, cost, latency, deployment, and failure modes. They make accountability visible again. This video explains why the “frontier advantage” is mostly narrative:Independent evaluations now show that open-source AI models can handle most everyday business tasks—summarizing documents, answering questions, drafting content, and internal analysis—at levels comparable to paid systems. The LMSYS Chatbot Arena, which runs blind human comparisons between models, consistently ranks open models close to top proprietary ones.Major consultancies now document why enterprises are switching: predictable costs, data control, and fewer legal and governance risks. McKinsey notes that open models reduce vendor lock-in and compliance exposure in regulated environments.Thanks for reading Design of AI: Strategies for Product Teams & Agencies! Subscribe for free to receive new posts and support my work.What Happens When “Freedom” Becomes an Excuse Not to Set Boundaries?We've confused freedom with capability. If a system can do something, we assume it should. That logic dissolves moral boundaries and replaces responsibility with abstraction: the model did it, the system allowed it.When no one owns the boundary, harm becomes an emergent property instead of a design failure.What If AI Doesn't Have to Be Owned by Corporations?We're going to experience a rise in AI experts challenging the expectations that Silicon Valley should control AI.What if AI doesn't need to be centralized, rented, or governed exclusively by corporate interests?On-device models and open ecosystems offer a different future—less extraction, fewer opaque incentives, and more meaningful choice.Follow Antoine Valot as him and Postcapitalist Design Club explore new ways of liberating AI.Are We Using AI for Anything That Actually Matters?Much of today's AI usage is performative productivity and ego padding that signals relevance while eroding self-trust. We're outsourcing thinking we are still capable of doing ourselves.AI should amplify judgment and creativity. Use this insanely powerful technology to make you achieve greater outcomes, not deliver a higher amount of subpar work to the world.If We Know the Risks Now, Why Are We Still Acting Surprised?The paper “The AI Model Risk Catalog” removes the last excuse.Failure modes are documented. Harms are mapped. Blind spots are known.Continuing to deploy without contingency planning is no longer innovation—it's negligence. If a team can't explain how its system fails safely, who intervenes, and what happens next, it isn't ready for real-world use.If Guardrails Don't Work, What Actually Protects Us?Every AI model and product is at risk of a major attack and exploit.AI systems are structurally vulnerable. The reason we haven't seen a catastrophic failure yet isn't safety—it's limited adoption and permissions.Guardrails fail under pressure. Policies collapse at scale. The only real protection is limiting blast radius: constraining autonomy and refusing to grant authority systems can't safely hold.Why Should Teams Decide Before They Build?The Decision-Forcing AI Business Case Canvas from Unhyped is essential for planning how to leverage AI in your products.Before discussing capabilities, teams must answer:* Who is accountable when this fails?* What judgment must remain human?* What harms are unacceptable—even if the system works?This canvas offers alignment on vision, responsibility, and impact isn't bureaucracy.It's baseline design discipline.Consider the TradeoffsThe conversation with Ovetta Sampson challenges a belief that shaped the last phase of AI adoption: that faster is always better, and that dependence on OpenAI, Google, or Anthropic is inevitable.That belief works during experimentation.It breaks the moment your product starts to matter.As teams scale, speed stops being the constraint. Trust, cost predictability, and accountability take its place. The question shifts from How fast can we ship? to What are we tying our business to—and what happens when it fails?One path optimizes for immediate momentum and simplicity. The other requires more upfront effort, but fundamentally changes where risk, data, and control live.This isn't a technical choice. It's a business one.As usage grows, externalized risk stops being abstract and starts showing up in margins, contracts, and customer trust.As that pressure builds, the impact becomes visible in the product experience itself.Latency creeps in. Costs compound quietly. Outputs vary in ways teams struggle to explain. What once felt powerful starts to feel fragile. Teams spend more time managing side effects than delivering value.At that point, you realize you didn't just choose a model.You chose a UX trajectory.Frontier models feel impressive early, but often lead to expensive, inconsistent experiences over time. Smaller, tuned models trade spectacle for reliability—and reliability is what users actually trust.Eventually, the conversation moves from UX to business fundamentals.Token pricing that felt negligible becomes material. Vendor updates change behavior you didn't choose. Security and compliance questions become harder to abstract away. You realize that outsourcing intelligence also outsourced leverage.This final image makes the tradeoff explicit. Paid frontier models buy speed and simplicity. Open or self-managed approaches buy independence, cost control, and long-term defensibility. Pretending these lead to the same outcomes is the mistake.This transition, from novelty to ownership, is exactly where Right AI Now is focused. Through her consultancy, Ovetta helps teams redesign AI decisions around outcomes that actually matter at scale: customer trust, data sovereignty, operational stability, and long-term value creation.These are also the themes we hear most consistently from the Design of AI audience. Founders and product leaders aren't asking for more tools—they're asking for clearer decisions. They want to know why AI products succeed and fail. We'll be going deeper on this shift throughout 2026, including a rebrand of the podcast, name and all.Improve Your AI ProductIf your organization is at the inflection point where AI needs to deliver real value without eroding trust, this is where I can help you. I've worked with teams at Microsoft, Spotify, and Mozilla to help leaders decide what to build, how to deliver value, and prioritize roadmaps. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit designofai.substack.com

Today's guest is Hayden Mitchell, Ph.D.  Hayden is a sports performance coach, educator, and researcher specializing in movement ecology and pedagogy, helping coaches design environments that support learning, resilience, self-actualization, and sustainable athletic performance through play and exploration. There is a great deal of conversation in sports performance around methods, including exercises, drills, systems, and models, but far less attention is given to coaching itself. Coaching methodology quietly shapes how athletes experience training, how they relate to challenge and failure, and ultimately how fully they are able to express themselves in performance. On the show today, Hayden speaks about exploring how coaching and physical education shape not just performance, but the whole human being. Hayden shares his path through sport, teaching, and doctoral work, including how life experiences changed his approach to leadership, control, and play. Together they discuss movement ecology, value orientations in coaching, such as mastery, learning process, self-actualization, social responsibility, and ecological integration, and why environment often matters as much as programming. The conversation highlights rhythm, joy, and exploration, along with practical ways coaches can use restraint, better questions, and playful constraints to help athletes own their development. Today's episode is brought to you by Hammer Strength. Use the code “justfly20” for 20% off any Lila Exogen wearable resistance training, including the popular Exogen Calf Sleeves. For this offer, head to Lilateam.com Use code “justfly10” for 10% off the Vert Trainer View more podcast episodes at the podcast homepage. (https://www.just-fly-sports.com/podcast-home/) Timestamps 0:00 – Hayden's coaching background 6:42 – Learning through experimentation 13:55 – Movement quality versus output 21:18 – Constraints based coaching 30:07 – Strength that transfers 39:50 – Variability and resilience 48:26 – Developing youth athletes 57:41 – Decision-making under fatigue 1:06:10 – Simplifying training programs 1:14:22 – Long term coaching philosophy Actionable Takeaways 6:42 – Learning through experimentation builds better coaches and athletes. Early coaching growth often comes from trying ideas, observing outcomes, and refining approaches. Allow room for trial and error in training rather than locking into rigid systems too early. Encourage athletes to feel and explore movement solutions instead of chasing perfect reps. Reflection after sessions helps clarify what actually transferred versus what just looked good. 13:55 – Movement quality creates the foundation for sustainable performance. Chasing outputs too early can hide inefficient movement strategies. Build positions, shapes, and rhythm before emphasizing max speed or max load. Use submaximal work to groove coordination and reduce compensation patterns. Improved movement quality often raises outputs without directly training them. 21:18 – Constraints guide learning better than constant verbal correction. Design drills that naturally guide athletes toward desired solutions. Reduce cue overload by letting the task do the teaching. Constraints promote adaptability instead of dependency on coaching feedback. This approach scales well in team settings with limited coaching bandwidth. 30:07 – Strength training should support movement, not replace it. Choose lifts that reinforce postures and force directions seen in sport. Avoid chasing strength numbers that disrupt rhythm or coordination. Use strength work to enhance confidence and robustness, not fatigue accumulation. Strong athletes still need to move well under dynamic conditions. 39:50 – Variability is a key driver of resilience. Expose athletes to multiple movement patterns and speeds. Avoid over standardizing drills to the point of robotic execution. Small variations build adaptability without sacrificing intent. Resilient athletes tolerate change better during competition. 48:26 – Youth athletes need exposure, not specialization. Prioritize broad skill development over early performance metrics. Multiple sports and movement environments improve long term ceilings. Avoid labeling young athletes too early based on temporary traits. Early diversity reduces burnout and overuse issues. 57:41 – Decision-making matters when athletes are tired. Fatigue reveals movement habits and decision quality. Train cognition alongside physical outputs when appropriate. Simple competitive games expose real world decision challenges. Performance under fatigue reflects true readiness. 1:06:10 – Simple programs executed well outperform complex plans done poorly. Clarity improves athlete buy in and consistency. Fewer exercises done with intent beat bloated sessions. Complexity should serve adaptation, not ego. Great programs are easy to repeat and sustain. 1:14:22 – Long term development requires patience and perspective. Short term gains should not compromise future potential. Progress is rarely linear, especially in young athletes. Coaching success is measured in years, not weeks. Build athletes you would want to train again in five years. Quotes from Hayden “Good movement solves a lot of problems before strength ever enters the conversation.” “When you design the environment well, you do not need to talk nearly as much.” “Outputs are easy to measure, but they are not always the most important thing.” “Variability is not chaos. It is preparation.” “Athletes who only know one solution struggle when conditions change.” “Young athletes do not need more specialization, they need more experiences.” “Strength should support expression, not restrict it.” “Simple does not mean easy. It means intentional.” “Fatigue exposes habits, not flaws.” “The goal is not just better athletes, but athletes who last.” About Hayden Mitchell Hayden Mitchell, PhD is a sports performance coach, educator, and researcher whose work sits at the intersection of movement ecology, pedagogy, and human development. He has coached and taught across a wide range of settings, from youth and collegiate sport to military, adaptive populations, and general fitness, working with ages 4 to 90. Hayden holds a doctorate in Human Performance and Sport Pedagogy and focuses on how environment, values, and teaching behaviors shape learning, resilience, and performance. His work emphasizes play, rhythm, and self-actualization, helping coaches and athletes move beyond rigid systems toward practices that develop both performance capacity and the whole human being.

In this episode of Tech Talks, Julian Dibbell is joined by partner Brian Nolan and associate Megan Fitzgerald to unpack how companies can protect AI assets and outputs using IP strategies. The conversation maps the key protectable components of AI—algorithms and code, trained models and parameters, proprietary datasets, and outputs—and evaluates the strengths and limits of trade secrets, copyrights, patents, and contracts. They highlight why trade secrets are particularly powerful for AI while probing emerging "improper means" issues like scraping, prompt injection, and ToS violations. They also survey evolving copyright law on human authorship and fair use in training, and discuss patent inventorship guidance and eligibility trends, before closing with practical contracting approaches to allocate data rights, output ownership, and IP strategy. Show Notes: 00:02 Introduction to Protecting AI Assets and Outputs 02:00 Protectable AI Assets: Algorithms, Models, Data, Outputs 06:07 IP Toolkit: Trade Secrets, Copyright, Patents, Contracts 09:55 "Improper means" in AI: Scraping, Prompts, ToS 12:44 Using Copyright to Protect AI 16:21 Copyright: Human Authorship, Code Protection, Fair Use 21:27 Patents: Inventorship Guidance, Eligibility, Open Issues 29:58 Contracts: Data Rights, Output Ownership, Strategy

Join the All In Mastermind: 100 Men, committed to their goals - https://www.muscleintelligence.com/apply/   Truth is, 98% of people set goals… and end the year in the same place. In this solo episode, Ben breaks down the 12 "Power Moves" framework he uses with elite founders, executives, and high performers to create repeatable wins in 2026. You'll learn why measuring inputs is a trap, how to build an outcomes-based scorecard for your body and life, and the real definition of success: intelligence + agency. Ben also explains the Mission–Map–Mentor model, the 4 resiliencies that determine follow-through (body, mind, stress, energy), and the 12 power moves that created exponential change in his own life. If you want a year that actually moves the needle, start here. 5 Bullet Points: Why "more information" can keep you stuck The scorecard that turns goals into outcomes Intelligence vs agency: the real success equation The 4 resiliencies that predict follow-through 12 Power Moves to build momentum fast   Whenever you're ready... here are 3 ways we can help you look, feel and perform at your best:   1. Grab a free copy of 1 of our BRAND NEW Peak Performance Protocols. This is for high performers looking to 10x their training and nutrition results by becoming 10x more effective. Click here - https://go.muscleintelligence.com/high-performance-executive-report/   2. Join the Muscle Intelligence Community and connect with other men like you who want to uplevel their health and fitness. It's our new Facebook group where I coach members live, share what's working with my private clients and announce tickets to my upcoming trainings and events. Click here - https://www.muscleintelligence.com/community   3. Read the Newsletter Join 200,000 men in their prime, reading our weekly newsletter: http://muscleintelligence.com/newsletter   Time Stamps: 00:00 Introduction to Power Moves 01:25 Curating Inputs and Outputs 02:17 Scorecards for Body, Health, and Wealth 06:13 Understanding Agency and Intelligence 20:12 The Importance of Strong Humans 22:17 Taking Personal Responsibility 22:36 Power Moves for Success 24:29 Creating a Scoreboard 27:33 Mastering Your Environment 28:35 Mastering Your Morning 30:32 Nurturing Family and Marriage 31:54 Connecting to a Higher Purpose 34:15 Becoming an A Player 37:48 Final Thoughts and Mentorship Invitation

This time Ned, Adam and Laura talk targets - and why the third Cycling and Walking Investment Strategy (CWIS3) needs outputs, not simply outcomes. They are joined by the CEO of the Walk, Wheel, Cycle Trust (formerly Sustrans), Xavier Brice, who knows all about strategies, and delivering active transport networks.The government recently ended a consultation on CWIS3 but, frustratingly, the proposals lacked any investment or much strategy. There were no SMART targets, or any outputs, i.e. routes; simply the unachievable outcome that by 2035 walking, wheeling and cycling will be "a safe, easy and accessible option for everyone". Road Investment Strategies, by contrast, focus heavily on routes and infrastructure, so why do we treat walking, wheeling and cycling differently?Xavier Brice has been CEO of the Walk Wheel Cycle Trust since 2016. In 2007 Brice led the development of a new walking and cycling strategy for London, with Transport for London.This month Adam, Laura and Xavier Brice coordinated an open letter to the Secretary of State supporting a better CWIS3. That letter was signed by more than 50 organisations across health, active travel and beyond. It asked that central government maps a true national network of routes by 2030, and sets targets to deliver that network to a proper, accessible standard by 2050.You can read the letter here: https://bsky.app/profile/adamtranter.bsky.social/post/3m7fv3vhyks2rThe letter was covered in the Guardian by Peter Walker: https://www.theguardian.com/politics/2025/dec/12/drivers-cyclists-transport-policy-conservatives-culture-wars-road-safety Shortly after that, Walker interviewed transport minister, Lilian Greenwood, about the importance of 'creating a system that works for everyone': https://www.theguardian.com/politics/2025/dec/12/drivers-cyclists-transport-policy-conservatives-culture-wars-road-safetyLaura's Freedom of Information requests to English local authorities found just 2 per cent had used legal powers to purchase land - something that's done routinely for roads https://substack.com/home/post/p-178788505And her article on CWIS3: https://lauralaker.substack.com/p/a-cycling-and-walking-strategy-walksThe Walk, Wheel Cycle Trust has been improving the National Cycle Network (NCN). In 2023/24 1.7km of an off-road muddy track connecting the residential area of Newton, in West Doncaster, to Danum retail park, was widened (on NCN62), with seven barriers removed or redesigned, along with improved wayfinding and signage. Estimated annual usage rose by 196% according to the Walk, Wheel Cycle Trust, from 150,000 trips in 2022 to 450,000 in 2024. Pedestrian and cycling trips increased by 191% and 192% respectively, while other users increased by 270%. Another path improvement project in Redcar and Cleveland saw ten barriers removed on NCN1 and NCN68. Wheelchair user trips increased four-fold, from 200 to 800, with 100% of disabled users saying they now use the route as the most convenient option.For ad-free listening, behind-the-scenes and bonus content and to help support the podcast - head to (https://www.patreon.com/StreetsAheadPodcast). We'll even send you some stickers! We're also on Bluesky and welcome your feedback on our episode: https://bsky.app/profile/podstreetsahead.bsky.social Hosted on Acast. See acast.com/privacy for more information.

How do volunteer leaders move from being seen as “extra hands” to strategic drivers of mission success? In this episode of the Volunteer Nation Podcast, Tobi Johnson is joined by Chris Wade and Matthew Cobble, co-hosts of the Time for Impact Podcast in the UK, for a practical and thought-provoking conversation about building influence through impact. Together, they explore why volunteering needs to be reframed as community participation and talent, not just unpaid labor and how leaders of volunteers can use data, stories, and strategic thinking to elevate their role inside organizations. This episode goes beyond counting hours or outputs and dives into how volunteer engagement directly contributes to outcomes, organizational strategy, and long-term change. Full show notes: 193. Building Influence with Impact with Chris Wade and Matthew Cobble Building Influence - Episode Highlights [00:31] - Introducing Special Guests: Chris Wade and Matthew Cobble [01:12] - Building Influence with Impact [01:57] - Meet Chris Wade: A Leader in Volunteerism [03:58] - Meet Matthew Cobble: A Journey in Volunteer Engagement [07:42] - The Importance of Volunteerism in Today's World [12:42] - Volunteers as a Strategic Asset [14:10] - Measuring Impact and Building Influence [24:12] - Challenges and Solutions in Volunteer Leadership [31:15] - Hypotheses and Program Design [32:18] - Vision Week and Volunteer Planning [33:06] - Shifting Mindsets on Volunteerism [34:12] - Strategic Planning and Data Utilization [36:13] - Design Thinking in Volunteer Management [37:39] - Collaborative Data Collection [40:32] - Practical How-Tos for Volunteer Impact [42:46] - Measuring Volunteer Impact [53:44] - Collecting Evidence and Surveys Helpful Links VolunteerPro Impact Lab  2025 Volunteer Management Progress Report – The Recruitment Edition  Time for Impact Podcast, Tobi Johnson on the Challenging, Brave Journey of Volunteer Leadership  Volunteer Nation Episode #175: Outputs vs Outcomes: Why Counting Hours Isn't Enough Info on Lewin's Force Field Analysis Info on Balanced Scorecard for Nonprofits  Info on the Double Diamond Design Process  Info on the Outcomes Star   Thanks for listening to this episode of the Volunteer Nation podcast. If you enjoyed it, please be sure to subscribe, rate, and review so we can reach more people like you who want to improve the impact of their good cause. For more tips and notes from the show, check us out at TobiJohnson.com. For any comments or questions, email us at WeCare@VolPro.net.

In this solo episode, Sam ties a bow on our ratios series — walking us through how to go from a rookie to an expert. Many recruiters are flying blind — not knowing what exactly is working, or why. Not knowing, definitively, what it takes to make a great placement. Sam breaks down the ratios of candidates needed for screening calls, interviews, and placements, contrasting the approaches of rookie and expert recruiters.Another takeaway? The importance of tracking recruitment metrics and creating a structured plan to achieve your goals effectively. You can't double-down on winning behaviors if you don't know what they are — so step one is to experiment, step two is to document, and step three is to create a feedback loop that ensures continued success. 

Every few years, the world of product management goes through a phase shift. When I started at Microsoft in the early 2000s, we shipped Office in boxes. Product cycles were long, engineering was expensive, and user research moved at the speed of snail mail. Fast forward a decade and the cloud era reset the speed at which we build, measure, and learn. Then mobile reshaped everything we thought we knew about attention, engagement, and distribution.Now we are standing at the edge of another shift. Not a small shift, but a tectonic one. Artificial intelligence is rewriting the rules of product creation, product discovery, product expectations, and product careers.To help make sense of this moment, I hosted a panel of world class product leaders on the Fireside PM podcast:• Rami Abu-Zahra, Amazon product leader across Kindle, Books, and Prime Video• Todd Beaupre, Product Director at YouTube leading Home and Recommendations• Joe Corkery, CEO and cofounder of Jaide Health • Tom Leung (me), Partner at Palo Alto Foundry• Lauren Nagel, VP Product at Mezmo• David Nydegger, Chief Product Officer at OvivaThese are leaders running massive consumer platforms, high stakes health tech, and fast moving developer tools. The conversation was rich, honest, and filled with specific examples. This post summarizes the discussion, adds my own reflections, and offers a practical guide for early and mid career PMs who want to stay relevant in a world where AI is redefining what great product management looks like.Table of Contents* What AI Cannot Do and Why PM Judgment Still Matters* The New AI Literacy: What PMs Must Know by 2026* Why Building AI Products Speeds Up Some Cycles and Slows Down Others* Whether the PM, Eng, UX Trifecta Still Stands* The Biggest Risks AI Introduces Into Product Development* Actionable Advice for Early and Mid Career PMs* My Takeaways and What Really Matters Going Forward* Closing Thoughts and Coaching Practice1. What AI Cannot Do and Why PM Judgment Still MattersWe opened the panel with a foundational question. As AI becomes more capable every quarter, what is left for humans to do. Where do PMs still add irreplaceable value. It is the question every PM secretly wonders.Todd put it simply: “At the end of the day, you have to make some judgment calls. We are not going to turn that over anytime soon.”This theme came up again and again. AI is phenomenal at synthesizing, drafting, exploring, and narrowing. But it does not have conviction. It does not have lived experience. It does not feel user pain. It does not carry responsibility.Joe from Jaide Health captured it perfectly when he said: “AI cannot feel the pain your users have. It can help meet their goals, but it will not get you that deep understanding.”There is still no replacement for sitting with a frustrated healthcare customer who cannot get their clinical data into your system, or a creator on YouTube who feels the algorithm is punishing their art, or a devops engineer staring at an RCA output that feels 20 percent off.Every PM knows this feeling: the moment when all signals point one way, but your gut tells you the data is incomplete or misleading. This is the craft that AI does not have.Why judgment becomes even more important in an AI worldDavid, who runs product at a regulated health company, said something incredibly important: “Knowing what great looks like becomes more essential, not less. The PM's that thrive in AI are the ones with great product sense.”This is counterintuitive for many. But when the operational work becomes automated, the differentiation shifts toward taste, intuition, sequencing, and prioritization.Lauren asked the million dollar question. “How are we going to train junior PMs if AI is doing the legwork. Who teaches them how to think.”This is a profound point. If AI closes the gap between junior and senior PMs in execution tasks, the difference will emerge almost entirely in judgment. Knowing how to probe user problems. Knowing when a feature is good enough. Knowing which tradeoffs matter. Knowing which flaw is fatal and which is cosmetic.AI is incredible at writing a PRD. AI is terrible at knowing whether the PRD is any good.Which means the future PM becomes more strategic, more intuitive, more customer obsessed, and more willing to make thoughtful bets under uncertainty.2. The New AI Literacy: What PMs Must Know by 2026I asked the panel what AI literacy actually means for PMs. Not the hype. Not the buzzwords. The real work.Instead of giving gimmicky answers, the discussion converged on a clear set of skills that PMs must master.Skill 1: Understanding context engineeringDavid laid this out clearly: “Knowing what LMS are good at and what they are not good at, and knowing how to give them the right context, has become a foundational PM skill.”Most PMs think prompt engineering is about clever phrasing. In reality, the future is about context engineering. Feeding models the right data. Choosing the right constraints. Deciding what to ignore. Curating inputs that shape outputs in reliable ways.Context engineering is to AI product development what Figma was to collaborative design. If you cannot do it, you are not going to be effective.Skill 2: Evals, evals, evalsRami said something that resonated with the entire panel: “Last year was all about prompts. This year is all about evals.”He is right.• How do you build a golden dataset.• How do you evaluate accuracy.• How do you detect drift.• How do you measure hallucination rates.• How do you combine UX evals with model evals.• How do you decide what good looks like.• How do you define safe versus unsafe boundaries.AI evaluation is now a core PM responsibility. Not exclusively. But PMs must understand what engineers are testing for, what failure modes exist, and how to design test sets that reflect the real world.Lauren said her PMs write evals side by side with engineering. That is where the world is going.Skill 3: Knowing when to trust AI output and when to override itTodd noted: “It is one thing to get an answer that sounds good. It is another thing to know if it is actually good.”This is the heart of the role. AI can produce strategic recommendations that look polished, structured, and wise. But the real question is whether they are grounded in reality, aligned with your constraints, and consistent with your product vision.A PM without the ability to tell real insight from confident nonsense will be replaced by someone who can.Skill 4: Understanding the physics of model changesThis one surprised many people, but it was a recurring point.Rami noted: “When you upgrade a model, the outputs can be totally different. The evals start failing. The experience shifts.”PMs must understand:• Models get deprecated• Models drift• Model updates can break well tuned prompts• API pricing has real COGS implications• Latency varies• Context windows vary• Some tasks need agents, some need RAG, some need a small finetuned modelThis is product work now. The PM of 2026 must know these constraints as well as a PM of the cloud era understood database limits or API rate limits.Skill 5: How to construct AI powered prototypes in hours, not weeksIt now takes one afternoon to build something meaningful. Zero code required. Prompt, test, refine. Whether you use Replit, Cursor, Vercel, or sandboxed agents, the speed is shocking.But this makes taste and problem selection even more important. The future PM must be able to quickly validate whether a concept is worth building beyond the demo stage.3. Why Building AI Products Speeds Up Some Cycles and Slows Down OthersThis part of the conversation was fascinating because people expected AI to accelerate everything. The panel had a very different view.Fast: Prototyping and concept validationLauren described how her teams can build working versions of an AI powered Root Cause Analysis feature in days, test it with customers, and get directional feedback immediately.“You can think bigger because the cost of trying things is much lower,” she said.For founders, early PMs, and anyone validating hypotheses, this is liberating. You can test ten ideas in a week. That used to take a quarter.Slow: Productionizing AI featuresThe surprising part is that shipping the V1 of an AI feature is slower than most expect.Joe noted: “You can get prototypes instantly. But turning that into a real product that works reliably is still hard.”Why. Because:• You need evals.• You need monitoring.• You need guardrails.• You need safety reviews.• You need deterministic parts of the workflow.• You need to manage COGS.• You need to design fallbacks.• You need to handle unpredictable inputs.• You need to think about hallucination risk.• You need new UI surfaces for non deterministic outputs.Lauren said bluntly: “Vibe coding is fast. Moving that vibe code to production is still a four month process.”This should be printed on a poster in every AI startup office.Very Slow: Iterating on AI powered featuresAnother counterintuitive point. Many teams ship a great V1 but struggle to improve it significantly afterward.David said their nutrition AI feature launched well but: “We struggled really hard to make it better. Each iteration was easy to try but difficult to improve in a meaningful way.”Why is iteration so difficult.Because model improvements may not translate directly into UX improvements. Users need consistency. Drift creates churn. Small changes in context or prompts can cause large changes in behavior.Teams are learning a hard truth: AI powered features do not behave like typical deterministic product flows. They require new iteration muscles that most orgs do not yet have.4. The PM, Eng, UX Trifecta in the AI EraI asked whether the classic PM, Eng, UX triad is still the right model. The audience was expecting disagreement. The panel was surprisingly aligned.The trifecta is not going anywhereRami put it simply: “We still need experts in all three domains to raise the bar.”Joe added: “AI makes it possible for PMs to do more technical work. But it does not replace engineering. Same for design.”AI blurs the edges of the roles, but it does not collapse them. In fact, each role becomes more valuable because the work becomes more abstract.• PMs focus on judgment, sequencing, evaluation, and customer centric problem framing• Engineers focus on agents, systems, architecture, guardrails, latency, and reliability• Designers focus on dynamic UX, non deterministic UX patterns, and new affordances for AI outputsWhat does changeAI makes the PM-Eng relationship more intense. The backbone of AI features is a combination of model orchestration, evaluation, prompting, and context curation. PMs must be tighter than ever with engineering to design these systems.David noted that his teams focus more on individual talents. Some PMs are great at context engineering. Some designers excel at polishing AI generated layouts. Some engineers are brilliant at prompt chaining. AI reveals strengths quickly.The trifecta remains. The skill distribution within it evolves.5. The Biggest Risks AI Introduces Into Product DevelopmentWhen we asked what scares PMs most about AI, the conversation became blunt and honest. Risk 1: Loss of user trustLauren warned: “If people keep shipping low quality AI features, user trust in AI erodes. And then your good AI product suffers from the skepticism.”This is very real. Many early AI features across industries are low quality, gimmicky, or unreliable. Users quickly learn to distrust these experiences.Which means PMs must resist the pressure to ship before the feature is ready.Risk 2: Skill atrophyTodd shared a story that hit home for many PMs. “Junior folks just want to plug in the prompt and take whatever the AI gives them. That is a recipe for having no job later.”PMs who outsource their thinking to AI will lose their judgment. Judgment cannot be regained easily.This is the silent career killer.Risk 3: Safety hazards in sensitive domainsDavid was direct: “If we have one unsafe output, we have to shut the feature off. We cannot afford even small mistakes.”In healthcare, finance, education, and legal industries, the tolerance for error is near zero. AI must be monitored relentlessly. Human in the loop systems are mandatory. The cycles are slower but the stakes are higher.Risk 4: The high bar for AI compared to humansJoe said something I have thought about for years: “AI is held to a much higher standard than human decision making. Humans make mistakes constantly, but we forgive them. AI makes one mistake and it is unacceptable.”This slows adoption in certain industries and creates unrealistic expectations.Risk 5: Model deprecation and instabilityRami described a real problem AI PMs face: “Models get deprecated faster than they get replaced. The next model is not always GA. Outputs change. Prompts break.”This creates product instability that PMs must anticipate and design around.Risk 6: Differentiation becomes hardI shared this perspective because I see so many early stage startups struggle with it.If your whole product is a wrapper around an LLM, competitors will copy you in a week. The real differentiation will not come from using AI. It will come from how deeply you understand the customer, how you integrate AI with proprietary data, and how you create durable workflows.6. Actionable Advice for Early and Mid Career PMsThis was one of my favorite parts of the panel because the advice was humble, practical, and immediately useful.A. Develop deep user empathy. This will become your biggest differentiator.Lauren said it clearly: “Maintain your empathy. Understand the pain your user really has.”AI makes execution cheap. It makes insight valuable.If you can articulate user pain precisely.If you can differentiate surface friction from underlying need.If you can see around corners.If you can prototype solutions and test them in hours.If you can connect dots between what AI can do and what users need.You will thrive.Tactical steps:• Sit in on customer support calls every week.• Watch 10 user sessions for every feature you own.• Talk to customers until patterns emerge.• Ask “why” five times in every conversation.• Maintain a user pain log and update it constantly.B. Become great at context engineeringThis will matter as much as SQL mattered ten years ago.Action steps:• Practice writing prompts with structured context blocks.• Build a library of prompts that work for your product.• Study how adding, removing, or reordering context changes output.• Learn RAG patterns.• Learn when structured data beats embeddings.• Learn when smaller local models outperform big ones.C. Learn eval frameworksThis is non negotiable.You need to know:• Precision vs recall tradeoffs• How to build golden datasets• How to design scenario based evals for UX• How to test for hallucination• How to monitor drift• How to set quality thresholds• How to build dashboards that reflect real world input distributionsYou do not need to write the code.You do need to define the eval strategy.D. Strengthen your product senseYou cannot outsource product taste.Todd said it best: “Imagine asking AI to generate 20 percent growth for you. It will not tell you what great looks like.”To strengthen your product sense:• Review the best products weekly.• Take screenshots of great UX patterns.• Map user flows from apps you admire.• Break products down into primitives.• Ask yourself why a product decision works.• Predict what great would look like before you design it.The PMs who thrive will be the ones who can recognize magic when they see it.E. Stay curiousRami's closing advice was simple and perfect: “Stay curious. Keep learning. It never gets old.”AI changes monthly. The PM who is excited by new ideas will outperform the PM who clings to old patterns.Practical habits:• Read one AI research paper summary each week.• Follow evaluation and model updates from major vendors.• Build at least one small AI prototype a month.• Join AI PM communities.• Teach juniors what you learn. Nothing accelerates mastery faster.F. Embrace velocity and side projectsTodd said that some of his biggest career breakthroughs came from solving problems on the side.This is more true now than ever.If you have an idea, you can build an MVP over a weekend. If it solves a real problem, someone will notice.G. Stay close to engineeringNot because you need to code, but because AI features require tighter PM engineering collaboration.Learn enough to be dangerous:• How embeddings work• How vector stores behave• What latency tradeoffs exist• How agents chain tasks• How model versioning works• How context limits shape UX• Why some prompts blow up API costsIf you can speak this language, you will earn trust and accelerate cycles.H. Understand the business deeplyJoe's advice was timeless: “Know who pays you and how much they pay. Solve real problems and know the business model.”PMs who understand unit economics, COGS, pricing, and funnel dynamics will stand out.7. Tom's Takeaways and What Really Matters Going ForwardI ended the recording by sharing what I personally believe after moderating this discussion and working closely with a variety of AI teams over the past 2 years.Judgment becomes the most valuable PM skillAs AI gets better at analysis, synthesis, and execution, your value shifts to:• Choosing the right problem• Sequencing decisions• Making 55 45 calls• Understanding user pain• Making tradeoffs• Deciding when good is good enough• Defining success• Communicating vision• Influencing the orgAgents can write specs.LLMs can produce strategies.But only humans can choose the right one and commit.Learning speed becomes a competitive advantageI said this on the panel and I believe it more every month.Because of AI, you now have:• Infinite coaches• Infinite mentors• Infinite experts• Infinite documentation• Infinite learning loopsA PM who learns slowly will not survive the next decade. Curiosity, empathy, and velocity will separate great from goodMany panelists said versions of this. The common pattern was:• Understand users deeply• Combine multiple tools creatively• Move quickly• Learn constantlyThe future rewards generalists with taste, speed, and emotional intelligence.Differentiation requires going beyond wrapper appsThis is one of my biggest concerns for early stage founders. If your entire product is a wrapper around a model, you are vulnerable.Durable value will come from:• Proprietary data• Proprietary workflows• Deep domain insight• Organizational trust• Distribution advantage• Safety and reliability• Integration with existing systemsAI is a component, not a moat.8. Closing ThoughtsHosting this panel made me more optimistic about the future of product management. Not because AI will not change the job. It already has. But because the fundamental craft remains alive.Product management has always been about understanding people, making decisions with incomplete information, telling compelling stories, and guiding teams through ambiguity and being right often.AI accelerates the craft. It amplifies the best PMs and exposes the weak ones. It rewards curiosity, empathy, velocity, and judgment.If you want tailored support on your PM career, leadership journey, or executive path, I offer 1 on 1 career, executive, and product coaching at tomleungcoaching.com.OK team. Let's ship greatness. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit firesidepm.substack.com

Most sellers dabble with AI prompts, but the results often come back generic, off-brand, or flat-out wrong. That's because effective prompts aren't about luck , they follow a recipe.In Part 2 of our 3-part AI for Sales series, we took you beyond definitions and into practical frameworks.Learn how to structure prompts step by step, spot the difference between weak and strong instructions, and apply proven prompt patterns that generate reliable outputs every time.This is a hands-on and tactical session, with good vs. bad examples, a starter prompt library, and troubleshooting tips you can apply instantly. By the end, you'll know how to turn everyday sales tasks into plug-and-play workflows anyone on your team can run.You'll Learn:The proven recipe for writing prompts that generate high-quality outputsGood vs. bad examples of prompts for emails, recaps, and researchA starter library of plug-and-play prompts sellers can use immediatelyThe Speaker:Jed MahrleIf you want to catch The Daily Sales Show live, join hereFollow Sell Better to get the latest actionable tactics from sales pros at the top of their gameExplore our YouTube ChannelThank you to our sponsors: Aligned and Winn.AI

Work with a DDS coach: https://datadrivenstrength.com/coaching/0:02:14 - Zac's Work on Individual Response Variation (and his conclusion)0:08:35 - Key takeaway: programming around constraints, not different principles0:15:18 - New podcast approach: Building on their own work (Research & Coaching)0:19:03 - Experience vs weak Scientific Evidence0:29:17 - Coaching Systems Review: The 5 Core Values in the training process0:37:48 - The value of Training Skill for long-term success0:44:36 - Listener Question: How to interpret years of Training Data0:52:43 - Coaching as the Application of Principles within the athlete's Constraints0:59:11 - Key Inputs and Outputs to evaluate Strength & Hypertrophy programming1:17:50 - The 3 Big Unanswered Questions in Training Science (Fatigue, Creatine, Growth Limit)

Rob Wilson is a performance educator with over twenty years of experience helping people build durable, high-functioning bodies and minds. He joins us on The Ready State Podcast to unpack the uncomfortable truth about performance: it's not free. In this powerful conversation with Kelly and Juliet Starrett, Rob dives into the real price of pushing your limits, why sleep is non-negotiable, and how to reframe “selfish” self-care as the foundation for showing up better in every area of life. Together, they tackle burnout, aging, and what it takes to sustain health and high output in a world that rewards constant hustle.What You'll Learn in This EpisodeThe three waves of fitness, and why Rob Wilson's book represents the vanguard of the third wave.The problem with the democratization of health metrics like Heart Rate Variability (HRV) if you don't know how to interpret the data or take action.The story behind the "Check Engine Light" metaphor, which helps high performers prioritize what to address and what to ignore.Why the phrase "self-care" often fails with service-oriented and high-performing individuals and the analogy used instead.The "Cobra Effect" or Goodhart's Law, and how chasing a metric like a high HRV can lead to misleading and useless outcomes.How to stop the "medical cascade" and apply an experimental framework (test/retest) to chronic, nagging pain and everyday health issues.The true cost of high performance and the crucial need for a "cost mitigation strategy" to avoid burnout.Why context matters more than perfect protocols, and how to create a personal longevity dashboard for continuous adaptation.For more info, follow Rob on Instagram and definitely pick up a copy of his new book, Check Engine Light: Tuning Your Body and Mind to Achieve Performance Longevity.Key Highlights: (00:00) - Intro(00:48) - Check Engine Light Book Overview(06:49) - Check Engine Light Metaphor Explained(14:06) - Importance of Check Engine Light for Everyone(17:49) - Inputs and Outputs in Life(20:37) - One Size Fits All Approach: Myth or Reality?(25:25) - Identifying What Matters Most to You(27:28) - Performance Costs: Understanding Trade-offs(29:27) - Recommended Supplements for Health(32:55) - LMNT: Importance of Hydration Explained(37:10) - Resistance: Creativity's Universal Challenge(39:48) - Becoming Reasonable: A Personal Journey(45:18) - Heart Rate Variability (HRV): Benefits Explained(47:15) - Using Tracking Devices Mindfully(49:57) - The Cobra Effect: Understanding Consequences(57:55) - Setting Up Environments for Success(1:00:20) - Changes Since Writing the Book(1:03:14) - What's Next for Rob: Future Plans(1:04:30) - Finding Rob: Where to ConnectSponsorsThis episode of The Ready State Podcast is brought to you by LMNT and Momentous. 

What does it mean to focus on outcomes over outputs? In this podcast hosted by iDonate VP of Product Nacho Andrade, LucidLink Chief Product Officer Richard Yu will be speaking on driving product strategy through outcome-focused innovation. Richard shares his unique approach to product management, blending technical insight with commercial strategy to create transformative solutions that solve real-world problems.

Most founders think they have a “people problem” or a “process problem.” The truth? It's almost always both. Without clear systems, people get frustrated. Without empowered people, processes get ignored. And now with AI and automation accelerating, the stakes are higher than ever.In this episode of Founder Talk, Ryan Weiss, founder of EPS Optics, shares how he helps companies streamline workstreams, align teams, and prepare for a future where AI is rapidly reshaping jobs. From diagnosing broken processes to balancing structure with creativity, Ryan explains why the companies that win are the ones who combine people and process to create real impact.We dive into how poor order entry created billing chaos at one client, why “healthy conflict” is essential for accountability, and what happens when you let blind spots hold your business back. Ryan also shares his journey from building a lawn care company at 15, to living in the Philippines and building outsourcing teams, to writing Optics, his Amazon bestselling book on process and perception.You'll learn:✅ Why most business frustrations come from missing processes or ignored systems✅ How to balance creativity with structure in your team✅ Why AI will replace many jobs and how to adapt before it happens✅ The SIPOC framework (Suppliers, Inputs, Process, Outputs, Customers) that transforms workflows✅ Why the future belongs to leaders who impact people, not just profitsIf you've been searching “how to fix broken processes,” “people vs. process in business,” or “how AI will impact jobs,” this episode gives you the no-fluff truth.Connect with RyanGuest LinkedIn: https://www.linkedin.com/in/ryancweiss/Guest Website: https://learnmore.epsoptics.com/If you are a B2B company that wants to build your own in-house content team instead of outsourcing your content to a marketing agency, we may be a fit for you! Everything you see in our podcast and content is a result of a scrappy, nimble, internal content team along with an AI-powered content systems and process. Check out pricing and services here: ⁠https://impaxs.com⁠Timecodes00:00 Introduction and Name Pronunciation00:12 German Heritage and Pronunciation Variations00:54 The Importance of Process in Business03:49 Balancing Process and Creativity06:59 Diagnosing and Solving Process Issues19:03 The Role of External Experts31:32 Living and Working in the Philippines34:16 The Future of Customer Service and AI34:38 AI Replacing Jobs: The Future of Work35:22 Streamlining Business Operations36:01 Preparing for Automation and AI37:40 Impact of AI on Computer Science Careers38:39 Adapting to Technological Changes43:36 The Importance of Mindset in Career Evolution49:14 Writing a Book: Process and Benefits56:09 Building a Business and Making an Impact59:55 Connecting and Growing Through Relationships

Udhay Durai, Executive Director of Data Platform and Engineering at Evolus, joins the show to unpack his journey from consulting to leading enterprise data teams. He shares how the high-pressure, quick-delivery mindset from consulting can be a secret weapon in a corporate setting, and what changes when you shift from delivering outputs to owning long-term outcomes. From navigating different types of pressure to building sustainable systems that scale, Udhay offers candid insights for anyone considering a similar transition.Key Takeaways• The consulting mindset of speed and adaptability can be a major advantage in enterprise roles when paired with long-term thinking• Pressure exists in both consulting and full-time roles, but the nature of that pressure—and how you manage it—differs greatly• Consultants focus on outputs, while enterprise leaders are measured on outcomes that stand the test of time• Generalist experience across domains can complement deep subject matter experts in a corporate team• Bringing incremental change and a “flywheel” approach from consulting can accelerate enterprise delivery without sacrificing reliabilityTimestamped Highlights01:34 — Why quick wins and stakeholder empathy are essential in consulting03:28 — How the pressure changes when you own the platform instead of just delivering a project05:32 — Outputs vs outcomes and why the shift matters in enterprise leadership09:48 — Turning generalist consulting experience into an asset in a full-time role11:43 — The biggest mindset and skill gaps to address when making the switch13:42 — Adapting consulting habits for long-term success in product companiesQuote of the Episode“Pressure is there in both consulting and enterprise. The difference is in consulting you deliver outputs—enterprise leaders deliver outcomes.”Resources MentionedUdhay Durai on LinkedIn — https://www.linkedin.com/in/udhay-duraiCall to ActionIf this episode gave you new perspective on career transitions, share it with a colleague or friend who's considering a similar move. Follow the show for more real-world tech leadership conversations.

In this episode of the Volunteer Nation Podcast, Tobi Johnson dives into why it's time to move beyond traditional volunteer metrics like hours logged and retention rates. She unpacks the difference between outputs and outcomes, sharing practical examples and actionable strategies for measuring and showcasing the true impact volunteers have on communities and organizations.  Tobi explains how tracking outcomes can inspire volunteers, improve program quality, strengthen recruitment messaging, guide strategic decisions, and demonstrate real value to stakeholders. You'll also learn how to create visual graphics to communicate these outcomes! Full show notes: 175. Outputs vs Outcomes: Why Counting Hours Isn't Enough Outputs vs Outcomes - Episode Highlights [02:47] - Proving Volunteer Impact [03:37] - Key Metrics to Track [06:14] - The Importance of Outcome Metrics [09:41] - Communicating Impact Effectively [13:58] - Why Volunteer Outcomes Matter [13:58] - Why Volunteer Outcomes Are Essential [18:36] - Communicating Volunteer Impact [22:24] - Improving Program Quality with Outcomes [25:40] - Strengthening Recruitment Messaging [28:42] - Supporting Strategic Decision Making [31:53] - Creating Visual Impact Graphics  Helpful Links Volunteer Management Progress Report  VolunteerPro Impact Lab Volunteer Nation Episode #135: How to Use Video Storytelling to Connect with Aaron Walton and Emmanuel LeGrair  Thanks for listening to this episode of the Volunteer Nation podcast. If you enjoyed it, please be sure to subscribe, rate, and review so we can reach more people like you who want to improve the impact of their good cause. For more tips and notes from the show, check us out at TobiJohnson.com. For any comments or questions, email us at WeCare@VolPro.net.

“Agile isn't about following the rules. It's about delivering real value together.” In this episode, host Rebecca Kalogeris speaks with Jenny Martin, seasoned facilitator, coach, and creator of the OOPSI framework. While Agile has transformed software development over the past two decades, many organizations struggle to scale it effectively or to see the benefits they were promised. Jenny explains why so many teams fall into the trap of focusing on ceremonies and tools instead of the principles that actually drive results, like collaboration, value delivery, and rapid feedback. She introduces OOPSI—short for Outcomes, Outputs, Process, Scenarios, Inputs—a lightweight, non-prescriptive framework that helps teams break down complex problems, align on value, and accelerate delivery. Jenny shares how OOPSI can resolve common Agile pitfalls like “water-scrum-fall,” where work still flows through waterfall-style handoffs despite sprint-based development. If your teams don't understand user stories, are unclear on priorities, or struggle to collaborate across functions, this conversation offers a practical path to restoring focus, alignment, and energy in Agile. For show notes and more resources, visit: pragmaticinstitute.com/resources/podcasts Pragmatic Institute is the global leader in Product, Data, and Design training and certification programs for working professionals. Learn more at pragmaticinstitute.com.

On this episode of the Self-Publishing News Podcast, Dan Holloway explains how the European Accessibility Act will impact authors selling into the EU, with practical tips on accessible formats and why EPUB matters more than ever. He also covers a major U.S. court ruling against Anthropic for scraping copyrighted books—while still declaring its AI-generated content fair use. Sponsors Self-Publishing News is proudly sponsored by Bookvault. Sell high-quality, print-on-demand books directly to readers worldwide and earn maximum royalties selling directly. Automate fulfillment and create stunning special editions with BookvaultBespoke. Visit Bookvault.app today for an instant quote. Self-Publishing News is also sponsored by book cover design company Miblart. They offer unlimited revisions, take no deposit to start work and you pay only when you love the final result. Get a book cover that will become your number-one marketing tool. Find more author advice, tips, and tools at our Self-publishing Author Advice Center, with a huge archive of nearly 2,000 blog posts and a handy search box to find key info on the topic you need. And, if you haven't already, we invite you to join our organization and become a self-publishing ally. About the Host Dan Holloway is a novelist, poet, and spoken word artist. He is the MC of the performance arts show The New Libertines, He competed at the National Poetry Slam final at the Royal Albert Hall. His latest collection, The Transparency of Sutures, is available on Kindle.