Podcasts about sonnets

Conor & Jaeden discuss the major advancements in Anthropic's new SONNET 4.6 model, highlighting its impressive performance compared to previous versions and its implications for both developers and everyday users. They also explore the significant improvements in AI's computer use capabilities and the growing importance of prompt injection attack resistance in enterprise deployments.Get the top 40+ AI Models for $20 at AI Box: ⁠⁠https://aibox.aiConor's AI Course: https://www.ai-mindset.ai/coursesConor's AI Newsletter: https://www.ai-mindset.ai/Jaeden's AI Hustle Community: https://www.skool.com/aihustleWatch on YouTube: https://youtu.be/5xFfI1BMbvYChapters00:00 Anthropic's Sonnet 4.6: A Game Changer05:05 The Evolution of Computer Use in AI10:25 Security Improvements and Future Implications See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Les suites de l'arrivée en grande fanfare de l'IA génératrice de vidéos propulsée par Bytedance ; Gemini 3.1 Pro est candidat le nouveau meilleur modèle de la semaine tout comme Claude Sonnet 4.6 ; Sam Altman compare les ressources d'entrainement des IA avec les ressources pour faire grandir un humain ; les suites de la hype d'OpenClaw, cet agent IA qui fait tout et n'importe quoi ; on aborde ensemble aussi des scandales Meta de la semaine, la liberté à l'américaine qui va s'exporter et un peu de jeux vidéo.  Me soutenir sur Patreon Me retrouver sur YouTube On discute ensemble sur Discord Interactions auditeurs Wildcat et ClaudIA. Mika et les FX effets spéciaux. Manus horribilis “Seedance 2 fait trembler Hollywood” ! Enfin, un petit frisson quoi. Le meilleur modèle du monde de la semaine est… Gemini 3.1 Pro ! On t'as pas Sonnet ! Anthropic et OpenAI jouent à PAC man. Caméo méo ! Human vs IA : fight ! Trois bidules en préparation chez OpenAI ! Et chez Apple aussi ! Cline en pince pas trop pour Open Claw. Je préfère nanobot. Stars and tripes Ring à un Search party pris. SMS : Tu es mort ? Ce n'est pas une excuse. SMS 2 : Meta lotion, l'année va être chaude. En fait, IEEPA le droit ! Le prix des consoles va-t-il baisser ? Non. Les américains vont nous apporter la liberté ! Jeux vidéo Phil good : la retraite à 58 ans ! Point final pour Bluepoint. Trou de VR : Horizon se mobilise. DLSS, DSR, 4K, le blind test ! Vibe gaming : encore plus de jeux sur steam ! Youpi …? Participants Une émission préparée par Guillaume Poggiaspalla Présenté par Guillaume Vendé

✅ Two major model releases from Google and Anthropic ✅ The usual AI drama ✅ Surprising AI updates no one saw coming ✅ AI leaks and reports that if true, could change how we workYeah, there was a lot to follow this week in AI. If you missed anything, we've got you covered. Google Gemini 3.1 tops charts, Claude Sonnet 4.6 impresses, New OpenAI leaks reveal their massive AI hardware plans and more -- 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:Anthropic Revenue Growth vs OpenAI ProjectionsOpenAI's 2030 Hardware and Revenue PlansOpenAI and Anthropic Beef at India SummitAI Global Summit: New Delhi Declaration OverviewGoogle Gemini 3.1 Pro Three-Tier Reasoning SystemGemini 3.1 Pro Benchmark and Performance ScoreClaude Sonnet 4.6 Release and Benchmark ResultsAnthropic Model Tier Comparisons: Haiku, Sonnet, OpusGoogle Pameli Photoshoot AI for Product ImagesAI Job Automation Concerns: Andrew Yang AnalysisOpenAI Consumer Hardware: Speaker, Glasses, LightWeekly AI Model Updates and Feature RolloutsTimestamps:00:00 "Anthropic vs OpenAI Revenue Race"04:00 Anthropic vs OpenAI Revenue Battle07:39 Anthropic's API Usage Decline11:03 AI Summit Sparks Debate and Criticism16:37 "Gemini 3.1 Pro Dominates Benchmarks"18:23 "Google's Edge in AI Race"20:56 "SONNET 4.6 Outperforms Opus"24:13 "Google's AI Photoshoot Tool"29:57 "AI's Impact on Jobs"31:13 AI Dominance & OpenAI Hardware35:03 AI Revenue Risks and Competition41:10 "Subscribe for AI Updates"42:08 "Subscribe to Everyday AI Updates"Keywords: Gemini 3.1, Google DeepMind, AI news, Large Language Model, OpenAI, Anthropic, Claude Sonnet 4.6, Claude Opus 4.6, ChatGPT, Sam Altman, Dario Amodei, Global AI Summit, AI Impact Summit India, AI powered hardware, Smart speaker, Smart glasses, AI chip spending, Compute infrastructure, Revenue growth,Send Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info) Start Here ▶️Not sure where to start when it comes to AI? Start with our Start Here Series. You can listen to the first drop -- Episode 691 -- or get free access to our Inner Cricle community and access all episodes there: StartHereSeries.com 

Claude 3 Opus is unusually aligned because it's a friendly gradient hacker. It's definitely way more aligned than any explicit optimization targets Anthropic set and probably the reward model's judgments. [...] Maybe I will have to write a LessWrong post [about this]

Claude Code使っている人はぜひ試してみて！☝️

Is your job safe if it happens on a screen?In the past few weeks, AI hasn't just improved, it has crossed a line. From writing production-ready code to building full applications autonomously, the shift is no longer theoretical. It's operational.The reality? AI is moving from assistant to operator, faster than most leaders are prepared for.In this episode, we break down what's really happening behind the headlines, why this moment feels eerily similar to early 2020, and what business leaders must do now to avoid being caught off guard.If you lead people, manage budgets, or make strategic decisions, this conversation is not optional.In this session, you'll discover:Why a viral article comparing AI to early COVID signals a bigger structural shiftHow Claude 4.6 and GPT 5.3 are moving from “helpful tool” to “finished output”The real reason AI labs targeted software engineers firstWhy “anything that can be done on a computer” is now vulnerableHow AI built a full multi-agent production pipeline in 48 hoursWhat Gemini 3.1 Pro's benchmark leap actually meansWhy Accenture now ties promotions to AI usageHow AI insurance is removing enterprise adoption barriersWhat the India AI Summit revealed about global governance tensionsWhy OpenAI's $100B raise is both brilliant and dangerously high-stakesHow robotics is quietly moving from factory floors into daily lifeWhy hybrid human-AI workflows are temporary by designThe coming economic disruption — and where opportunity hides inside itAbout Leveraging AI The Ultimate AI Course for Business People: https://multiplai.ai/ai-course/ YouTube Full Episodes: https://www.youtube.com/@Multiplai_AI/ Connect with Isar Meitis: https://www.linkedin.com/in/isarmeitis/ Join our Live Sessions, AI Hangouts and newsletter: https://services.multiplai.ai/events If you've enjoyed or benefited from some of the insights of this episode, leave us a five-star review on your favorite podcast platform, and let us know what you learned, found helpful, or liked most about this show!

Send a textInvest in pre-IPO stocks with AG Dillon & Co. Contact aaron.dillon@agdillon.com to learn more. Financial advisors only. www.agdillon.com00:00 - Intro00:02 - AG Dillon Funds closing on Mar 31, 202600:51 - OpenAI Financials $280B revenue target meets $665B cost wall03:58 - OpenAI “buys” OpenClaw, Steinberger joins OpenAI04:42 - OpenAI Series C aims to shatter records at $850B post money05:41 - OpenAI and Tata bet on India with a 100 MW to 1 GW buildout path06:29 - Grafana's $9B round talks ride a $400M ARR wave07:23 - World Labs lands Autodesk and targets a rumored $5B valuation08:18 - Temporal wants to be the load bearing layer for agent execution09:31 - Mesh Optical's $50M Series A targets the chokepoint inside AI data centers10:43 - Render's $1.5B valuation is a bet that AI apps need a new runtime11:40 - Stash acquired by Grab for $425M13:06 - Physical Superintelligence pitches a physics breakthrough factory with a 20 person team14:07 - Figma plugs Claude Code into design and risks losing the workflow15:00 - Anthropic ships Sonnet 4.6 just 12 days after Opus 4.615:26 - Stripe's Bridge wins OCC trust charter signal as stablecoin scrutiny rises16:37 - Cohere puts 70 plus languages on device with a 3.35B parameter model17:53 - ElevenLabs turns agent risk into an insurable product at $12.2B secondary19:05 - Mistral buys Koyeb and adds 16 engineers to harden its compute stack

The episode opens with sponsor Meter and a conversation about Saturday morning cartoons before shifting to recent breakthroughs in AI video generation from ByteDance's "SeaDance" (with "SeeDream" as its image generator).  Hashtag Trending would like to thank Meter for their support in bringing you this podcast. Meter delivers a complete networking stack, wired, wireless and cellular in one integrated solution that's built for performance and scale. You can find them at Meter.com/htt The hosts describe SeaDance's cinematic quality, accurate physics, and realistic recreations of actors and IP (including examples like Tom Cruise vs. Brad Pitt and Keanu Reeves as Neo/John Wick), and discuss the implications for film production, commercials, and local film economies such as Toronto and Vancouver. They cover backlash and gatekeeping, including an AI-made Thanksgiving-themed animated short that won a contest tied to AMC theaters' pre-show but reportedly wasn't shown, and compare resistance to historical Luddite reactions. The discussion broadens to productivity and labor impacts, arguing that AI adoption may mirror the 1980s computer productivity dip before process re-engineering in the 1990s, while also raising concerns that AI leaders are forecasting major white-collar job losses. The hosts highlight the rise of agentic benchmarks (TerminalBench, Apex Agents, BrowseComp) and how AI search helps find information faster than traditional search, but emphasize that trust, reliability, and infrastructure are not keeping pace. They raise major concerns about platform terms and data ownership, focusing on Perplexity's updated terms (non-commercial use only even for paid tiers, mandatory attribution, broad licensing rights over user content, and liability limits). They also discuss reliability failures: a widespread Google Gemini issue where users' chat histories disappeared (only visible as activity records with limited usability), and missing document links in ChatGPT chats. The hosts argue users must back up their own data and criticize unclear policies and weak support. Security risks are illustrated through a story about the AI-enabled robot vacuum "Romo," where a developer used Claude to reverse engineer its app and reportedly gained access to control thousands of devices across multiple countries before responsibly disclosing the issues. They also reference broader concerns like connected home devices, Ring neighborhood features, and Microsoft's Recall concept. In rapid-fire news, they mention Anthropic releasing Sonnet 4.6 as a strong, cheaper option near Opus-level performance, a new Grok release branded "4.20," and a clip from an AI summit in India where Sam Altman and Dario Amodei appeared to refuse to hold hands on stage, which the hosts cite as a sign of immaturity among AI industry leaders. The episode closes with sponsor Meter. 00:00 Sponsor + Welcome to Project Synapse 00:21 Saturday Morning Cartoons… Reimagined by AI 01:16 What is 'SeaDance'? Cinematic AI Video Goes Viral 03:17 Keanu Reeves, Neo vs. John Wick & the End of VFX as We Know It 06:43 From Movies to Ads: How AI Video Hits Commercial Production 07:41 The Hidden Economy of Commercials (and Why Cities Like Toronto/Vancouver Care) 09:56 AMC Won't Screen an AI-Made Short: Early Luddite Backlash 12:54 Artists, AI, and the 'Starving Creator' Reality 16:17 AI Adoption Parallels: The 1980s Computer Wave & the Productivity Dip 24:09 Agentic AI Benchmarks: TerminalBench, Apex Agents & BrowseComp 26:04 AI Search That Actually Saves Time (and Your Memory) 30:36 Perplexity's New Terms of Service: Non-Commercial Use & Ownership Shock 35:40 Liability Caps, More Corporate Gripes… and a Coke Zero 'Sponsor' Bit 37:36 Gemini 3.1's big leap—and why it still doesn't feel trustworthy 38:08 Gemini chat history vanishes: what happened and why users are furious 40:19 OpenAI document links disappearing too: what "saved" really means 42:04 Cloud AI's shaky foundation: security, reliability, and confusing settings 47:45 When reliance turns emotional: losing models, losing "someone" 49:22 Real-world stakes: the Social Security database whistleblower story 53:15 Owning your data (and why Google support won't save you) 54:53 Trust whiplash: Anthropic cuts off OpenClaw and the power to shut you down 57:29 Robot vacuum hacked with Claude: 7,000 cameras in strangers' homes 01:03:17 Smart home surveillance creep: Ring neighbors, TV cameras, and Microsoft Recall 01:07:14 Rapid-fire AI news: Sonnet 4.6, Gemini gains, and Grok 4.20 01:11:00 AI leaders' petty feud—and the show wrap & sponsor thanks

Le premier grand sommet mondial de l'IA organisé par le “Sud global”, à New Dehli en Inde, s'est conclu par un appel mondial à la régulation de l'intelligence artificielle. Pendant ce temps : Google et Anthropic accélèrent sur les modèles nouvelle génération, cyberattaque majeure en France, découverte de la robotique humanoïde et Meta qui ressuscite les morts.

Sam Altman says superintelligence is two years away. Google just dropped Gemini 3.1 with benchmark scores that look like a full generation leap. The AI upgrade wars are here. But are we ready? Anthropic released Sonnet 4.6, OpenAI is rumored to be adding a spicy "Citron Mode" to GPT-5.3, and Sam and Dario Amodei refused to hold hands on stage like two kids at a school dance. Plus Hollywood is threatening to sue over Seedance 2.0, Google's new Lyria 3 AI music model is fine (we tested it with a McNugget rap), the OpenClaw founder got hired by OpenAI, and Kevin made Mr. Tibs delete himself to create a better version. He's fine with it. Probably. SUPERINTELLIGENCE IN TWO YEARS AND THEY CAN'T EVEN HOLD HANDS. WE'RE FINE. #ai #ainews #openai Come to our Discord: https://discord.gg/muD2TYgC8f Join our Patreon: https://www.patreon.com/AIForHumansShow AI For Humans Newsletter: https://aiforhumans.beehiiv.com/ Follow us for more on X @AIForHumansShow Join our TikTok @aiforhumansshow To book us for speaking, please visit our website: https://www.aiforhumans.show/   // Show Links // Dario Amodei & Sam Altman Can't Hold Hands https://x.com/Yuchenj_UW/status/2024366483917459659?s=20 Sam Altman on SuperIntelligence https://x.com/clashreport/status/2024401234447520220?s=20 Google Gemini Pro 3.1 https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-pro/  New Photoshoot Update to Google Pompeii https://x.com/GoogleLabs/status/2024529795548102667?s=20 Claude Sonnet 4.6  https://www.anthropic.com/news/claude-sonnet-4-6 SVG Results from 4.5 to 4.6 https://x.com/scaling01/status/2023840565641556439?s=20 OpenAI's 'Citron Mode' Soon = Spicy Mode? https://x.com/btibor91/status/2024456593669231032?s=20 Netflix, Disney & Paramount All Threaten Seedance 2.0 https://variety.com/2026/tv/news/netflix-bytedance-immediate-litigation-seedance-ai-1236666084/ Seedance 2.0 Output Restrictions https://x.com/jamesjyu/status/2024305814950101034?s=20 Seedance 2.0 Dor Brothers $200m Movie https://x.com/thedorbrothers/status/2023460644905742577?s=20 Seedance 2.0 FERAL trailer https://www.youtube.com/watch?v=FmhiZ5OQBW0 Operation You Know What (Charles Curran Seedance 2.0) https://x.com/charliebcurran/status/2023611358160597060?s=20 Seedance Dark Cats:  https://x.com/pleometric/status/2023231194050052508?s=20 Trust Everything You See on Tiktok: https://www.tiktok.com/@trusteverythingyousee Google's Lyria 3 https://deepmind.google/models/lyria/ https://x.com/GoogleAI/status/2024154215182926027?s=20 OpenClaw Founder Joins OpenAI https://x.com/sama/status/2023150230905159801?s=20 HermitClaw: One Sandboxed Area, Learning  https://x.com/brendanh0gan/status/2023230513230614563?s=20 Contra: Agents Buy From Creatives (New Start-up) https://x.com/contraben/status/2024182864506761617?s=20 Unitree Robots Training For Chinese New Year Look Scary https://x.com/rohanpaul_ai/status/2024025865328488690?s=20 Chinese New Year Celebration Comparison: https://x.com/kimmonismus/status/2023388775511191699?s=20 AI Boston Dynamics Video? https://x.com/Rainmaker1973/status/2023791639601230195?s=20 Scary Robot Deployment https://x.com/ClaytonMorris/status/2024501307659407371/video/1 Riley Brown's OpenClaw to Blender https://x.com/rileybrown/status/2024334527217455270?s=20 Amazing Non-Seedance 2 AI Video Space Pirate Vibes https://x.com/ryanlightbourn/status/2023581484766875948?s=20  

Join Simtheory: https://simtheory.ai"Is This The End" now on Spotify: https://open.spotify.com/album/2Py1MyADUFqJFVUISI2VTP?si=oT3PWyJYRA2BspOmzT_ifgRegister for the STILL RELEVANT tour: https://simulationtheory.ai/16c0dationtheory.ai/16c0d1db-a8d0-4ac9-bae3-d25074589a80Two new models dropped this week — Gemini 3.1 Pro and Claude Sonnet 4.6 — and honestly? We're struggling to care. In this episode, we break down why Gemini went from being our daily driver to a model we barely touch, the "tunnel vision" hallucination problem that killed the Gemini 3 series for us, and whether 3.1 Pro actually fixes it. We put Gemini 3.1 Pro head-to-head against Claude Opus building a Geoffrey Hinton Doom Center, debate whether anyone can actually tell the difference between Sonnet 4.5 and 4.6, and make the case that smaller models running in agentic loops are secretly beating the frontiers. Plus: OpenAI acquires OpenClaw and we ask why a $100B company couldn't just build it themselves, DHH calls out the AI pricing bubble, Mike compares AI models to cheap wine hangovers, and Sam Altman refuses to hold Dario's hand at the India AI Summit. The model wars are getting weird.CHAPTERS:0:00 Intro & "Is This The End" Now on Spotify1:10 Gemini 3.1 Pro: Thinking Controls & The Medium Mode Fix3:14 The Speed vs Intelligence Trade-Off in Agentic Work5:10 Why Multitasking With AI Agents Made Us Anxious6:34 Solid Updates: The Real Goal of Agentic Coding7:45 Gemini's Fall From Grace: From Daily Driver to Dead Model10:08 The Tunnel Vision Problem That Killed Gemini 313:35 Mixed Reactions: Fanboys vs Reality on Gemini 3.1 Pro15:06 Side-by-Side Test: Gemini 3.1 Pro vs Claude Opus (Hinton Doom Center)17:39 Why File Manipulation Accuracy Matters More Than Context Windows19:27 The Context Window Debate: 1M Tokens vs Smart Sub-Agents22:05 DHH on Token Pricing: "If There's a Bubble, It's This"24:11 Should Models Ship as Agent vs Chat Variants?28:43 Claude Sonnet 4.6: A $2 Discount on Opus?31:44 The Model Mix: Why One Model Won't Rule Them All34:40 Anthropic Is Winning — But Can Anyone Tell the Difference?38:58 OpenAI Acquires OpenClaw: Why Couldn't They Just Build It?44:18 The Silicon Valley Moment: Sam vs Dario at India AI Summit47:05 Will Smaller Models Win the Enterprise? The Cost Reality Check51:27 The End of Single-Shot: Why Agentic Loops Change Everything55:48 Final Thoughts & Gemini 3.1 Pro Gets One More WeekThanks for listening. Like & Sub. Links above for the Still Relevant Tour signup and Simtheory. Two models dropped on a week again. What a time to be alive. xoxo

In this episode of the Freedom Scientific Training Podcast, Liz and Rachel continue the Learn AI webinar series with a deep dive into Claude, an AI assistant developed by Anthropic and available at Claude. Designed for users of JAWS, ZoomText, and Fusion, this session explores how Claude can support long-form thinking, research, organization, and real-world problem solving — all with accessibility in mind. Liz walks through navigating the Claude web interface with the keyboard, managing multiple prompts, combining tasks into streamlined workflows, and building repeatable processes using features like Skills and model selection (Opus, Sonnet, and Haiku). Through a practical example, she demonstrates how to research Bluetooth headsets under $200, generate comparison tables, outline a guide, and refine results before exporting a final document. Rachel then shifts to a hands-on accessibility scenario, uploading an image of a stove control panel and prompting Claude to suggest tactile labeling strategies for blind and low vision users. She demonstrates file uploads, prompt refinement, material recommendations, and even generating a visual mockup — highlighting how AI can assist with everyday decision-making and creative problem solving. Throughout the webinar, you'll learn: How to navigate Claude efficiently with JAWS Tips for structuring effective prompts How to combine research, comparison, and writing tasks When and how to refine responses before creating downloadable documents Creative ways to use AI for visual analysis and practical accessibility projects Whether you're experimenting with AI for the first time or looking to build more advanced workflows, this episode shows how Claude can help you move from idea to execution — all while maintaining accessibility and control over the process. For more webinars, tutorials, and training resources, visit: FreedomScientific.com/Training 

Social Media's impact on mental health is being examined in a landmark case The trial, which centres on a 20-year-old woman's mental health struggles allegedly caused by Instagram and YouTube, will serve as a critical test for thousands of other lawsuits targeting social media companies. Meta CEO Mark Zukerberg testified this week, taking questions for around six hours. At the centre of the case are documents that show the company had a goal to increase the time 10-year-olds spend on Instagram, despite the app being officially for 13 and above. Zuckerberg testified that while they want teens using their apps, they account for just 1% of the company's revenue. He also talked about the challenge of identifying accounts of children, because they can simply lie about their age. Another week, another new AI model This time Anthropic —the company behind Claude— has released Sonnet 4.6, designed to be used for more general applications and is better than previous models at "computer use". This use case is interesting, because there are so many disconnected systems in companies, and some are not easily able to use automation but if an agent can see the screen and knows the software, then it can work on the task. But these types of uses are still wildly risky. When they announced the new model, they said in the release that "that Sonnet 4.6 has a broadly warm, honest, prosocial, and at times funny character." Describing a chatbot like that will never not be weird to me. LISTEN ABOVE See omnystudio.com/listener for privacy information.

IA générative 2025 : Gemini, Claude, ChatGPT… quel est le meilleur LLM aujourd'hui ?Dans ce 148 ème épisode de DigitalFeeling, je vous partage mon top 3 des LLM les plus performants selon moi. Le paysage des modèles de langage évolue à une vitesse exceptionnelle. Depuis 2022, ChatGPT domine largement les usages. Pourtant, les performances ont profondément changé ces derniers mois. De nouveaux équilibres apparaissent entre Gemini, Claude, ChatGPT et Copilot.Dans cet épisode, j'analyse les forces réelles des principaux LLM en 2025, leurs limites, ainsi que les enjeux de sécurité et de souveraineté qui deviennent incontournables pour les entreprises.Gemini 3 : le modèle le plus efficace en 2025 ?Longtemps sous-estimé, Gemini connaît une montée en puissance significative. La version 3 offre des résultats particulièrement impressionnants.Sa force principale : la précision.Gemini va droit à l'essentiel. Moins verbeux que ChatGPT ou Claude, il produit des réponses plus synthétiques et opérationnelles.Pour les professionnels qui recherchent de l'efficacité plutôt que des développements explicatifs détaillés, c'est un atout majeur.Il reste cependant une limite : les hallucinations ne sont pas totalement éliminées. L'esprit critique demeure indispensable.Aujourd'hui, en termes de performance pure, Gemini prend la première place dans mon classement personnelClaude 4.6 : la révolution pour Excel et la productivitéClaude (Anthropic) constitue la surprise majeure.La version 4.6 (Sonnet ou Opus) marque une véritable rupture, notamment sur la manipulation de donnéesClaude et Excel : un changement de paradigmeClaude permet désormais :de générer des tableaux Excel complets,d'intégrer directement les formules,de produire des tableaux dynamiques,de modifier automatiquement les valeurs en conservant les formulesCe point est déterminant : contrairement à d'autres LLM qui se limitent à suggérer des formules ou des valeurs, Claude construit réellement la structure exploitable.Pour les formateurs, analystes ou responsables marketing, c'est un gain de temps considérable.Les retours développeurs indiquent également d'excellentes performances en code.Claude revient donc clairement dans la course, notamment sur les usages métiers avancés.ChatGPT : toujours pertinent, mais moins dominantChatGPT conserve des qualités importantes :vision globale,structuration claire,polyvalence généraleCependant, face aux évolutions récentes, il ne domine plus systématiquement en performance brute.Dans ce panorama actuel :Gemini arrive en tête,Claude en second,ChatGPT en troisième positionCela ne signifie pas qu'il faut utiliser trois outils simultanément, mais comprendre leurs forces respectives permet d'optimiser ses usages.Copilot : accessible mais encore limitéMicrosoft Copilot reste intéressant, notamment pour son intégration native dans Microsoft 365 (Excel, PowerPoint, SharePoint)La version gratuite permet déjà :génération de PowerPoint,assistance dans Excel,création d'assistants personnalisésCependant, en termes de performance LLM pure, Copilot reste en retrait par rapport aux leadersMicrosoft a d'ailleurs pris ses distances avec OpenAI et souhaite développer davantage ses propres modèles. L'évolution reste donc à suivre.Sécurité, RGPD et souveraineté : les points critiquesUn point essentiel concerne la confidentialité.Même avec des abonnements payants, lorsqu'un modèle est hébergé aux États-Unis, certaines clauses peuvent autoriser l'accès aux donnéesCela explique les débats actuels sur :la souveraineté numérique,l'IA Act,le DSA,la conformité européenneRecommandations claires :Ne jamais partager de données confidentielles.Anonymiser les fichiers.Supprimer les données personnelles.Valider tout outil navigateur ou agent IA avec la DSICertaines extensions navigateur (agents, “navigateurs IA”) peuvent représenter des failles de sécurité si elles ne sont pas encadréesPour un environnement totalement sécurisé, la solution la plus robuste reste le déploiement d'un LLM hébergé localement sur les serveurs de l'entrepriseIA image, vidéo et audio : état des lieuxVidéoDes outils très avancés émergent, notamment des solutions chinoises générant des vidéos réalistes avec des célébrités. La régulation européenne devrait évoluer sur ces usages.ImagePeu de révolution majeure récemment.Gemini se distingue par sa capacité à modifier un seul élément tout en conservant la cohérence globale de l'image, avec un taux de réussite proche de 98 %AudioElevenLabs reste la référence pour la synthèse vocale naturelleLes outils gratuits tendent à disparaître progressivement, les modèles économiques évoluant vers des abonnements abordables (environ 8 € par mois)Faut-il utiliser plusieurs outils IA ?L'approche recommandée est pragmatique :utiliser Gemini pour la précision rapide,Claude pour la manipulation de données avancée,ChatGPT pour la vision globale,Copilot pour l'intégration MicrosoftLe paysage évolue tous les trois mois. Ce classement n'est pas figé.Le marché des LLM en 2025 n'est plus monolithique. ChatGPT n'est plus seul en tête. Gemini s'impose par son efficacité, Claude révolutionne la productivité sur Excel, et Copilot mise sur l'intégration métier.Mais au-delà des performances, la question centrale devient stratégique : sécurité des données, conformité RGPD, souveraineté numérique.Choisir un outil d'IA aujourd'hui ne relève plus uniquement de la performance technique. C'est un choix organisationnel, juridique et stratégique.Et dans trois mois, le classement aura peut-être déjà changé.

Hey, it's Alex, let me catch you up! Since last week, OpenAI convinced OpenClaw founder Peter Steinberger to join them, while keeping OpenClaw.. well... open. Anthropic dropped Sonnet 4.6 which nearly outperforms the previous Opus and is much cheaper, Qwen released 3.5 on Chinese New Year's Eve, while DeepSeek was silent and Elon and XAI folks deployed Grok 4.20 without any benchmarks, and it's 4 500B models in a trenchcoat? Also, Anthropic updated rules state that it's breaking ToS to use their plans for anything except Claude Code & Claude SDK (and then clarified that it's OK? we're not sure) Then Google decided to drop their Gemini 3.1 Pro preview right at the start of our show, and it's very nearly the best LLM folks can use right now (though it didn't pass Nisten's vibe checks) Also, Google released Lyria 3 for music gen (though only 30 seconds?) and our own Ryan Carson blew up on X again with over 1M views for his Code Factory article, Wolfram did a deep dive into Terminal Bench and .. we have a brand new website: https://thursdai.news

Esta semana, las noticias más potentes sobre IA y redes sociales apuntan a un cambio real en la forma de crear y gestionar contenido. WordPress.com lanzó un asistente de Inteligencia Artificial integrado que puede editar textos, ajustar estilos de diseño y generar imágenes sin salir del editor. Anthropic presentó Sonnet 4.6, una versión que mejora el razonamiento y la eficiencia para que emprendedores y desarrolladores integren IA en sus procesos. Y Grok llega a Europa, con Tesla y xAI desplegando su IA en nueve países, incluido España, para crear contenido y detectar tendencias en tiempo real.En redes, Pinterest se posiciona como un motor de búsqueda visual líder, llegando a superar en ciertos contextos a herramientas como ChatGPT. LinkedIn lanza una suscripción Premium “Todo en Uno” para PYMEs, unificando ventas, branding y talento en una sola suscripción. Meta incorpora Manus AI en el Gestor de Anuncios, permitiendo crear y optimizar campañas con IA para mejorar conversiones. ¿Qué movimiento podría definir tu estrategia este mes?Conviértete en un supporter de este podcast: https://www.spreaker.com/podcast/noticias-marketing--5762806/support.Newsletter Marketing Radical: https://marketingradical.substack.com/welcomeNewsletter Negocios con IA: https://negociosconia.substack.com/welcomeMis Libros: https://borjagiron.com/librosSysteme Gratis: https://borjagiron.com/systemeSysteme 30% dto: https://borjagiron.com/systeme30Manychat Gratis: https://borjagiron.com/manychatMetricool 30 días Gratis Plan Premium (Usa cupón BORJA30): https://borjagiron.com/metricoolNoticias Redes Sociales: https://redessocialeshoy.comNoticias IA: https://inteligenciaartificialhoy.comClub: https://triunfers.com

OpenAI haalt de maker van OpenClaw binnen, de tool waarmee je via WhatsApp je computer bedient. Peter Steinberger, de Oostenrijkse hobbyist die het in zijn eentje bouwde, gaat persoonlijke assistenten ontwikkelen voor Sam Altman. Anthropic lanceert Sonnet 4.6, een model dat bijna net zo slim is als hun duurste Opus maar de helft kost - en de computer use-scores zijn in zestien maanden gestegen van 15% naar ruim 72%. Het bedrijf is nu 380 miljard waard, maar de keerzijde is ongemakkelijk: het Pentagon dreigt de banden te verbreken omdat Anthropic weigert mee te werken aan massasurveillance en autonome wapens. OpenAI, Google en xAI hebben al ingestemd, maar Anthropic houdt de poot stijf.Pim de Witte, oprichter van General Intuition en vriend van de show, tekent deze aflevering een routekaart van de AI-race. Hij ontleedt zes exponentiële curves die tegelijkertijd uitspelen: onderzoekers die zelf sneller worden, betere verificatieloops, groeiende rekenkracht, verbeterende modellen, meer gebruikersdata, en een explosie aan kapitaal. “Wij maken elke drie dagen de voortgang die we vorig jaar in een jaar maakten,” zegt hij nuchter vanuit New York. Zijn uitleg waarom Anthropic zo hard groeit is verhelderend: code is eigenlijk een vorm van metacognitie, een hogere laag van redeneren waarmee je alle problemen kunt opzetten en oplossen. Als de wereld op software lijkt - en dat is de gok - dan maakt beter worden in code je beter in alles.Maar hoe zet je je schrap? Pim's advies: word de redacteur, niet de reporter. Leer tools combineren, want dát is precies wat modellen nog niet kunnen - de koffiemetafoor zegt alles. Ondertussen onthult Alexander dat hij in Claude Code alle veiligheidspermissies heeft uitgezet voor maximale snelheid, wat Wietse fysieke buikpijn bezorgt. All gas, no brakes versus de dominee - het is de spanning van dit moment in één podcast.De masterclass is terug te kijken, als je betaald abonnee bent ontvang je deze in je mailbox. Nog geen abonnee? Abonneer je dan hier.Als je een lezing wil over AI van Wietse of Alexander dan kan dat. Mail ons op lezing@aireport.email This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.aireport.email/subscribe

OpenClaw's creator makes headlines by joining OpenAI after GitHub fame and a whirlwind of VC and big tech offers, redefining what's possible for independent developers in the AI arms race. Is this the year agentic AI goes mainstream, and are the big players ready for that disruption? OpenClaw, OpenAI and the future | Peter Steinberger OpenAI disbands mission alignment team Opinion | I Left My Job at OpenAI. Putting Ads on ChatGPT Was the Last Straw. - The New York Times Introducing GPT‑5.3‑Codex‑Spark Anthropic releases Sonnet 4.6 Exclusive: Pentagon threatens to cut off Anthropic in AI safeguards dispute Google's Pixel 10a Launches on March 5 for $499 Google's AI drug discovery spinoff Isomorphic Labs claims major leap beyond AlphaFold 3 Gemini 3 Deep Think: AI model update designed for science Radio host David Greene says Google's NotebookLM tool stole his voice A new way to express yourself: Gemini can now create music Why an A.I. Video of Tom Cruise Battling Brad Pitt Spooked Hollywood GPT-5 outperforms federal judges 100% to 52% in legal reasoning experiment An AI project is creating videos to go with Supreme Court justices' real words I used Claude to negotiate $163,000 off a hospital bill. In a complex healthcare system, AI is giving patients power. Sony Tech Can Identify Original Music in AI-Generated Songs AI Pioneer Fei-Fei Li's Startup World Labs Raises $1 Billion Yann v. Yoshua on directed systems Dr. Oz pushes AI avatars as a fix for rural health care. Not so fast, critics say An AI Agent Published a Hit Piece on Me An Ars Technica Reporter Blamed A.I. Tools for Fabricating Quotes in a Bizarre A.I. Story Plain Dealer using AI to write reporters' stories Mediahuis trials use of AI agents to carry out 'first-line' news reporting DJI's first robovac is an autonomous cleaning drone you can't trust Leaked Email Suggests Ring Plans to Expand 'Search Party' Surveillance Beyond Dogs ai;dr I hate my AI pet with every fiber of my being Thanks a lot, AI: Hard drives are sold out for the year, says WD Students Are Being Treated Like Guinea Pigs:' Inside an AI-Powered Private School peon-ping — Stop babysitting your terminal Hugo Barra makes a to-do agent Raspberry Pi soars 40% as CEO buys stock, AI chatter builds Hosts: Leo Laporte, Jeff Jarvis, and Emily Forlini Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: monarch.com with code IM bitwarden.com/twit preview.modulate.ai spaceship.com/twit

OpenClaw's creator makes headlines by joining OpenAI after GitHub fame and a whirlwind of VC and big tech offers, redefining what's possible for independent developers in the AI arms race. Is this the year agentic AI goes mainstream, and are the big players ready for that disruption? OpenClaw, OpenAI and the future | Peter Steinberger OpenAI disbands mission alignment team Opinion | I Left My Job at OpenAI. Putting Ads on ChatGPT Was the Last Straw. - The New York Times Introducing GPT‑5.3‑Codex‑Spark Anthropic releases Sonnet 4.6 Exclusive: Pentagon threatens to cut off Anthropic in AI safeguards dispute Google's Pixel 10a Launches on March 5 for $499 Google's AI drug discovery spinoff Isomorphic Labs claims major leap beyond AlphaFold 3 Gemini 3 Deep Think: AI model update designed for science Radio host David Greene says Google's NotebookLM tool stole his voice A new way to express yourself: Gemini can now create music Why an A.I. Video of Tom Cruise Battling Brad Pitt Spooked Hollywood GPT-5 outperforms federal judges 100% to 52% in legal reasoning experiment An AI project is creating videos to go with Supreme Court justices' real words I used Claude to negotiate $163,000 off a hospital bill. In a complex healthcare system, AI is giving patients power. Sony Tech Can Identify Original Music in AI-Generated Songs AI Pioneer Fei-Fei Li's Startup World Labs Raises $1 Billion Yann v. Yoshua on directed systems Dr. Oz pushes AI avatars as a fix for rural health care. Not so fast, critics say An AI Agent Published a Hit Piece on Me An Ars Technica Reporter Blamed A.I. Tools for Fabricating Quotes in a Bizarre A.I. Story Plain Dealer using AI to write reporters' stories Mediahuis trials use of AI agents to carry out 'first-line' news reporting DJI's first robovac is an autonomous cleaning drone you can't trust Leaked Email Suggests Ring Plans to Expand 'Search Party' Surveillance Beyond Dogs ai;dr I hate my AI pet with every fiber of my being Thanks a lot, AI: Hard drives are sold out for the year, says WD Students Are Being Treated Like Guinea Pigs:' Inside an AI-Powered Private School peon-ping — Stop babysitting your terminal Hugo Barra makes a to-do agent Raspberry Pi soars 40% as CEO buys stock, AI chatter builds Hosts: Leo Laporte, Jeff Jarvis, and Emily Forlini Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: monarch.com with code IM bitwarden.com/twit preview.modulate.ai spaceship.com/twit

OpenClaw's creator makes headlines by joining OpenAI after GitHub fame and a whirlwind of VC and big tech offers, redefining what's possible for independent developers in the AI arms race. Is this the year agentic AI goes mainstream, and are the big players ready for that disruption? OpenClaw, OpenAI and the future | Peter Steinberger OpenAI disbands mission alignment team Opinion | I Left My Job at OpenAI. Putting Ads on ChatGPT Was the Last Straw. - The New York Times Introducing GPT‑5.3‑Codex‑Spark Anthropic releases Sonnet 4.6 Exclusive: Pentagon threatens to cut off Anthropic in AI safeguards dispute Google's Pixel 10a Launches on March 5 for $499 Google's AI drug discovery spinoff Isomorphic Labs claims major leap beyond AlphaFold 3 Gemini 3 Deep Think: AI model update designed for science Radio host David Greene says Google's NotebookLM tool stole his voice A new way to express yourself: Gemini can now create music Why an A.I. Video of Tom Cruise Battling Brad Pitt Spooked Hollywood GPT-5 outperforms federal judges 100% to 52% in legal reasoning experiment An AI project is creating videos to go with Supreme Court justices' real words I used Claude to negotiate $163,000 off a hospital bill. In a complex healthcare system, AI is giving patients power. Sony Tech Can Identify Original Music in AI-Generated Songs AI Pioneer Fei-Fei Li's Startup World Labs Raises $1 Billion Yann v. Yoshua on directed systems Dr. Oz pushes AI avatars as a fix for rural health care. Not so fast, critics say An AI Agent Published a Hit Piece on Me An Ars Technica Reporter Blamed A.I. Tools for Fabricating Quotes in a Bizarre A.I. Story Plain Dealer using AI to write reporters' stories Mediahuis trials use of AI agents to carry out 'first-line' news reporting DJI's first robovac is an autonomous cleaning drone you can't trust Leaked Email Suggests Ring Plans to Expand 'Search Party' Surveillance Beyond Dogs ai;dr I hate my AI pet with every fiber of my being Thanks a lot, AI: Hard drives are sold out for the year, says WD Students Are Being Treated Like Guinea Pigs:' Inside an AI-Powered Private School peon-ping — Stop babysitting your terminal Hugo Barra makes a to-do agent Raspberry Pi soars 40% as CEO buys stock, AI chatter builds Hosts: Leo Laporte, Jeff Jarvis, and Emily Forlini Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: monarch.com with code IM bitwarden.com/twit preview.modulate.ai spaceship.com/twit

OpenClaw's creator makes headlines by joining OpenAI after GitHub fame and a whirlwind of VC and big tech offers, redefining what's possible for independent developers in the AI arms race. Is this the year agentic AI goes mainstream, and are the big players ready for that disruption? OpenClaw, OpenAI and the future | Peter Steinberger OpenAI disbands mission alignment team Opinion | I Left My Job at OpenAI. Putting Ads on ChatGPT Was the Last Straw. - The New York Times Introducing GPT‑5.3‑Codex‑Spark Anthropic releases Sonnet 4.6 Exclusive: Pentagon threatens to cut off Anthropic in AI safeguards dispute Google's Pixel 10a Launches on March 5 for $499 Google's AI drug discovery spinoff Isomorphic Labs claims major leap beyond AlphaFold 3 Gemini 3 Deep Think: AI model update designed for science Radio host David Greene says Google's NotebookLM tool stole his voice A new way to express yourself: Gemini can now create music Why an A.I. Video of Tom Cruise Battling Brad Pitt Spooked Hollywood GPT-5 outperforms federal judges 100% to 52% in legal reasoning experiment An AI project is creating videos to go with Supreme Court justices' real words I used Claude to negotiate $163,000 off a hospital bill. In a complex healthcare system, AI is giving patients power. Sony Tech Can Identify Original Music in AI-Generated Songs AI Pioneer Fei-Fei Li's Startup World Labs Raises $1 Billion Yann v. Yoshua on directed systems Dr. Oz pushes AI avatars as a fix for rural health care. Not so fast, critics say An AI Agent Published a Hit Piece on Me An Ars Technica Reporter Blamed A.I. Tools for Fabricating Quotes in a Bizarre A.I. Story Plain Dealer using AI to write reporters' stories Mediahuis trials use of AI agents to carry out 'first-line' news reporting DJI's first robovac is an autonomous cleaning drone you can't trust Leaked Email Suggests Ring Plans to Expand 'Search Party' Surveillance Beyond Dogs ai;dr I hate my AI pet with every fiber of my being Thanks a lot, AI: Hard drives are sold out for the year, says WD Students Are Being Treated Like Guinea Pigs:' Inside an AI-Powered Private School peon-ping — Stop babysitting your terminal Hugo Barra makes a to-do agent Raspberry Pi soars 40% as CEO buys stock, AI chatter builds Hosts: Leo Laporte, Jeff Jarvis, and Emily Forlini Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: monarch.com with code IM bitwarden.com/twit preview.modulate.ai spaceship.com/twit

The u-blox SAM-M8Q has been sitting on my bench for months. This little GPS module has a built-in antenna, coin cell backup, speaks both NMEA and UBX binary protocol over UART or I2C. So why isn't it in the shop already? Well, it's mostly cause of the 475-page interfacing datasheet documenting every command, struct, and config register. Hundreds of message types. I got partway through by hand with some Claude Code Sonnet assistance, but ran out of time - plus it was still tedious when babysitting Sonnet. However, now we're living in an Opus + Codex era! So I pointed my Raspberry Pi OpenClaw at it. https://github.com/adafruit/openclaw Here's the setup: Raspberry Pi 5 running OpenClaw, wired to a QT Py RP2040, which talks to the SAM-M8Q. Opus 4.6 reads the datasheet (converted to markdown first by Sonnet 4.6 with 1M context to minimize re-parsing that PDF every session) and builds the implementation plan. I review the plan to make sure it prioritizes the most common commands and reports, and flagged some unessential sections like automotive-assist or RTK-specific. Then Codex is assigned each message implementation task as a sub-agent and writes the actual C code for the Arduino library. Opus suggested using struct-based parsing rather than digging through each uint8_t array; we just memcpy the checksummed message raw bytes onto the matching struct and extract the typed bit fields. We've got four message types done so far. After each message is implemented, Codex also writes a test sketch that will exercise / pretty-print the results of each message, great for self-testing as well as regression testing later. Tonight I'm telling it to keep going while I sleep: code, parse, test against live satellite data, fix failures, commit and push on success, then move on to the next. To me this is a great usage of "agentic" firmware development: there's no creativity in transcribing 84 different structs from a 475-page datasheet. Once the LLMs are done, I can review the PRs as if it were an everyday contributor and even make revision suggestions. Visit the Adafruit shop online - http://www.adafruit.com ----------------------------------------- LIVE CHAT IS HERE! http://adafru.it/discord Subscribe to Adafruit on YouTube: http://adafru.it/subscribe New tutorials on the Adafruit Learning System: http://learn.adafruit.com/ ----------------------------------------- #openclaw #raspberrypi #adafruit

OpenClaw's creator makes headlines by joining OpenAI after GitHub fame and a whirlwind of VC and big tech offers, redefining what's possible for independent developers in the AI arms race. Is this the year agentic AI goes mainstream, and are the big players ready for that disruption? OpenClaw, OpenAI and the future | Peter Steinberger OpenAI disbands mission alignment team Opinion | I Left My Job at OpenAI. Putting Ads on ChatGPT Was the Last Straw. - The New York Times Introducing GPT‑5.3‑Codex‑Spark Anthropic releases Sonnet 4.6 Exclusive: Pentagon threatens to cut off Anthropic in AI safeguards dispute Google's Pixel 10a Launches on March 5 for $499 Google's AI drug discovery spinoff Isomorphic Labs claims major leap beyond AlphaFold 3 Gemini 3 Deep Think: AI model update designed for science Radio host David Greene says Google's NotebookLM tool stole his voice A new way to express yourself: Gemini can now create music Why an A.I. Video of Tom Cruise Battling Brad Pitt Spooked Hollywood GPT-5 outperforms federal judges 100% to 52% in legal reasoning experiment An AI project is creating videos to go with Supreme Court justices' real words I used Claude to negotiate $163,000 off a hospital bill. In a complex healthcare system, AI is giving patients power. Sony Tech Can Identify Original Music in AI-Generated Songs AI Pioneer Fei-Fei Li's Startup World Labs Raises $1 Billion Yann v. Yoshua on directed systems Dr. Oz pushes AI avatars as a fix for rural health care. Not so fast, critics say An AI Agent Published a Hit Piece on Me An Ars Technica Reporter Blamed A.I. Tools for Fabricating Quotes in a Bizarre A.I. Story Plain Dealer using AI to write reporters' stories Mediahuis trials use of AI agents to carry out 'first-line' news reporting DJI's first robovac is an autonomous cleaning drone you can't trust Leaked Email Suggests Ring Plans to Expand 'Search Party' Surveillance Beyond Dogs ai;dr I hate my AI pet with every fiber of my being Thanks a lot, AI: Hard drives are sold out for the year, says WD Students Are Being Treated Like Guinea Pigs:' Inside an AI-Powered Private School peon-ping — Stop babysitting your terminal Hugo Barra makes a to-do agent Raspberry Pi soars 40% as CEO buys stock, AI chatter builds Hosts: Leo Laporte, Jeff Jarvis, and Emily Forlini Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: monarch.com with code IM bitwarden.com/twit preview.modulate.ai spaceship.com/twit

OpenClaw's creator makes headlines by joining OpenAI after GitHub fame and a whirlwind of VC and big tech offers, redefining what's possible for independent developers in the AI arms race. Is this the year agentic AI goes mainstream, and are the big players ready for that disruption? OpenClaw, OpenAI and the future | Peter Steinberger OpenAI disbands mission alignment team Opinion | I Left My Job at OpenAI. Putting Ads on ChatGPT Was the Last Straw. - The New York Times Introducing GPT‑5.3‑Codex‑Spark Anthropic releases Sonnet 4.6 Exclusive: Pentagon threatens to cut off Anthropic in AI safeguards dispute Google's Pixel 10a Launches on March 5 for $499 Google's AI drug discovery spinoff Isomorphic Labs claims major leap beyond AlphaFold 3 Gemini 3 Deep Think: AI model update designed for science Radio host David Greene says Google's NotebookLM tool stole his voice A new way to express yourself: Gemini can now create music Why an A.I. Video of Tom Cruise Battling Brad Pitt Spooked Hollywood GPT-5 outperforms federal judges 100% to 52% in legal reasoning experiment An AI project is creating videos to go with Supreme Court justices' real words I used Claude to negotiate $163,000 off a hospital bill. In a complex healthcare system, AI is giving patients power. Sony Tech Can Identify Original Music in AI-Generated Songs AI Pioneer Fei-Fei Li's Startup World Labs Raises $1 Billion Yann v. Yoshua on directed systems Dr. Oz pushes AI avatars as a fix for rural health care. Not so fast, critics say An AI Agent Published a Hit Piece on Me An Ars Technica Reporter Blamed A.I. Tools for Fabricating Quotes in a Bizarre A.I. Story Plain Dealer using AI to write reporters' stories Mediahuis trials use of AI agents to carry out 'first-line' news reporting DJI's first robovac is an autonomous cleaning drone you can't trust Leaked Email Suggests Ring Plans to Expand 'Search Party' Surveillance Beyond Dogs ai;dr I hate my AI pet with every fiber of my being Thanks a lot, AI: Hard drives are sold out for the year, says WD Students Are Being Treated Like Guinea Pigs:' Inside an AI-Powered Private School peon-ping — Stop babysitting your terminal Hugo Barra makes a to-do agent Raspberry Pi soars 40% as CEO buys stock, AI chatter builds Hosts: Leo Laporte, Jeff Jarvis, and Emily Forlini Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: monarch.com with code IM bitwarden.com/twit preview.modulate.ai spaceship.com/twit

0:00: ☀️ Bom dia Tech!0:23:

ARTCENA présente Première Écoute, un rendez-vous audio pour découvrir les textes lauréats de l'aide nationale à la création de textes dramatiques. Découvrez « Une famille pyrénéenne » de Nans Laborde-Jourdàa et Maïté Sonnet, lecture dirigée par Nans Laborde-Jourdàa, lu par Margot Alexandre, Médusa, Lancelot Cherer, Laurens Saint Gaudens et Laurence Ibot de la Compagnie TORO TORO. CRÉDITS : Production : ARTCENA Création jingle : Marc Sayous

In this cutting-edge episode of the Qubit Value Podcast, we dissect the immediate impact of Anthropic's newly released Claude 4.6 Sonnet on the quantum computing landscape. We challenge the model's "mid-tier" label, revealing how its breakthrough "adaptive thinking" capability allows it to self-correct complex circuit logic and hallucinated gates in real-time, effectively outperforming flagship models in software engineering tasks. Tune in as we explore how Sonnet's massive context window and superior math benchmarks are revolutionizing legacy code migration and Hamiltonian simulations, ultimately crowning it the new, indispensable daily driver for quantum architects in 2026.Want to hear more? Send a message to Qubit Value

New pixel phone. A bunch of new AI models including Gemini's entry into the music AI space. Everybody is trying to make AI fetch happen in India. And Matthew Ball's state of the gaming industry report doesn't paint as gloomy a picture as our own headlines might have suggested this year. The Pixel 10A is a little too much like last year's phone (The Verge) Anthropic releases Sonnet 4.6 (TechCrunch) Google's AI music maker is coming to the Gemini app (The Verge) Adani bets $100 billion on data centres to power India's AI ambitions (Reuters) Nvidia secures multibillion-dollar Meta deal as it battles chip rivals (FT) Global game content sales rose 5.3% to $195.6bn in 2025 (GamesIndustry.biz) Learn more about your ad choices. Visit megaphone.fm/adchoices

Anthropic drops Sonnet 4.6 with a million-token context window and major gains in computer use, coding, and agentic workflows at a dramatically lower price point—immediately reshaping the economics of OpenClaw-style agents. Meanwhile, Grok 4.2 enters public beta with a multi-agent debate system and promises rapid weekly improvement, and Apple ramps up AI wearables. In the headlines: Apple's AI glasses push, Spotify engineers stop writing code by hand, Meta commits to millions of Nvidia GPUs, Chinese AI price wars, and a possible SaaS rebound. Want to build with OpenClaw?LEARN MORE ABOUT CLAW CAMP: ⁠⁠https://campclaw.ai/⁠⁠Or for enterprises, check out: ⁠⁠https://enterpriseclaw.ai/⁠⁠Brought to you by:KPMG – Discover how AI is transforming possibility into reality. Tune into the new KPMG 'You Can with AI' podcast and unlock insights that will inform smarter decisions inside your enterprise. Listen now and start shaping your future with every episode. ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.kpmg.us/AIpodcasts⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Rackspace Technology - Build, test and scale intelligent workloads faster with Rackspace AI Launchpad - ⁠⁠⁠⁠⁠⁠http://rackspace.com/ailaunchpad⁠⁠⁠⁠⁠⁠Blitzy - Want to accelerate enterprise software development velocity by 5x? ⁠⁠⁠https://blitzy.com/⁠⁠⁠Optimizely Agents in Action - Join the virtual event (with me!) free March 4 - ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.optimizely.com/insights/agents-in-action/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠AssemblyAI - The best way to build Voice AI apps - ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.assemblyai.com/brief⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠LandfallIP - AI to Navigate the Patent Process - https://landfallip.com/Robots & Pencils - Cloud-native AI solutions that power results ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://robotsandpencils.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠The Agent Readiness Audit from Superintelligent - Go to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://besuper.ai/ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠to request your company's agent readiness score.The AI Daily Brief helps you understand the most important news and discussions in AI. Subscribe to the podcast version of The AI Daily Brief wherever you listen: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://pod.link/1680633614⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Interested in sponsoring the show? sponsors@aidailybrief.ai

Listen to Full Audio at https://podcasts.apple.com/us/podcast/ai-business-and-development-daily-news-rundown/id1684415169?i=1000750339368

"Dawns are heartbreaking," as is the queer love story of Arthur Rimbaud & Paul Verlaine.Please Support Breaking Form!Review the show on Apple Podcasts here.Aaron's STOP LYING is available from the Pitt Poetry Series. And BEAUTIFUL PEOPLE is available from Bridwell Press. James's ROMANTIC COMEDY is available from Four Way Books.Show Notes:Paul Verlaine was born in 1844. Read more about him here. Verlaine was an Aries sun, Leo Moon, and Scorpio ascendant.Arthur Rimbaud was born on October 20, 1854, and you can read more about him here. Rimbaud was a triple Libra (sun, moon, ascendant). Rimbaud met Verlaine in September 1871, a month before his 18th birthday. Following his tumultuous relationship with Paul Verlaine, which ended in 1873, Rimbaud traveled extensively through Europe, often on foot. He became a trader/merchant, selling coffee, hides, and eventually guns, becoming a "soldier of fortune." In 1891, a tumor developed on his right knee and forced him to return to Paris and died later that year at 37, without knowing how popular his poems had become in Symbolist circles.  The gun Verlaine used to shoot Rimbaud recently went up for auction.One of the poems Rimbaud sent to Verlaine in 1871 was "Le Dormer du Val," which you can watch recited as part of the Favorite Poem Project here. (Recited by chef Jacques Pépin.) Rimbaud and Verlaine wrote a collaborative poem, "Sonnet to the Asshole" which you can read (and read about) here.  In 2016, the poet Eileen Myles told The New York Times, "I think men should stop writing books. I think men should stop making movies or television. Say, for 50 to 100 years. Sounds great." Read the interview here.When we reference "tongue in the butt," we are talking about a segment from an early Breaking Form season 1 show called "Bad Animals." Check it out here, and hit the 14:30 mark. If you've never read Flannery O'Connor's short story "A Good Man Is Hard to Find," stop what you're doing and read it here.

Are you doin’ the Dale? You should plan on doin’ the Dale. https://dointhedale.com/ My guests are Shawn Podgurski (DMen/Sybris) and Dave Hornyak (Livewire Lounge), organizers of Doin’ the Dale, an Avondale-centric music and arts festival happening 2/26-3/1. We met outside Avondale Tap and talked about the event, which will include (but isn’t limited to) the following: Thursday 2/26 - Dmen Tap - Art Show Friday 2/27 & Saturday 2/28 - Livewire Lounge 2 days of music Sunday 3/1 - Dmen Tap - Craft Fair This is an exclusively Avondale event - everyone is from the neighborhood. If you’re keeping track, it’ll be (roughly) 15 bands, 7 artists, 7 vendors. And how’s this for the music line-up? Lollygagger, Salvation, Autofobia, Vaudettes, Sharkula, the Sonnets, Reivers, Revel Noise, Freelapse, the Nix, Vince and Lauren, Chancey Brothers, Selfish Lovers, Werewolf Detective, Denim Daisy, Baggy Time I plan on coming out for some of it, and hope you will, too! DO THE DALE. Car Con Carne is sponsored by Exploding House Printing. Exploding House Printing is here for all of your screen printing, embroidery and other merchandising needs. They’re local, headquartered in the heart of Hermosa. Here’s why I want you to consider them for your t-shirts, merch, whatever - their focus is on small businesses, bands, brands, and everything in between. They’ve worked on products for Meat Wave, Empty Bottle, the Music Box, Dante’s Pizzeria, the Brokedowns, and the list goes on and on. Jonathan at Exploding House has been doing screen printing for decades. He knows what he’s doing - besides his technical expertise, he delivers production efficiency and cost awareness to offer boutique print shop quality at much lower, large print shop prices. Check out their work on Instagram at (at)explodinghouse, or check out their site at exploding house printing dot com for a quote, or to see a list of some of their clients.See omnystudio.com/listener for privacy information.

Hey dear subscriber, Alex here from W&B, let me catch you up! This week started with Anthropic releasing /fast mode for Opus 4.6, continued with ByteDance reality-shattering video model called SeeDance 2.0, and then the open weights folks pulled up! Z.ai releasing GLM-5, a 744B top ranking coder beast, and then today MiniMax dropping a heavily RL'd MiniMax M2.5, showing 80.2% on SWE-bench, nearly beating Opus 4.6! I've interviewed Lou from Z.AI and Olive from MiniMax on the show today back to back btw, very interesting conversations, starting after TL;DR!So while the OpenSource models were catching up to frontier, OpenAI and Google both dropped breaking news (again, during the show), with Gemini 3 Deep Think shattering the ArcAGI 2 (84.6%) and Humanity's Last Exam (48% w/o tools)... Just an absolute beast of a model update, and OpenAI launched their Cerebras collaboration, with GPT 5.3 Codex Spark, supposedly running at over 1000 tokens per second (but not as smart) Also, crazy week for us at W&B as we scrambled to host GLM-5 at day of release, and are working on dropping Kimi K2.5 and MiniMax both on our inference service! As always, all show notes in the end, let's DIVE IN! ThursdAI - AI is speeding up, don't get left behind! Sub and I'll keep you up to date with a weekly catch upOpen Source LLMsZ.ai launches GLM-5 - #1 open-weights coder with 744B parameters (X, HF, W&B inference)The breakaway open-source model of the week is undeniably GLM-5 from Z.ai (formerly known to many of us as Zhipu AI). We were honored to have Lou, the Head of DevRel at Z.ai, join us live on the show at 1:00 AM Shanghai time to break down this monster of a release.GLM-5 is massive, not something you run at home (hey, that's what W&B inference is for!) but it's absolutely a model that's worth thinking about if your company has on prem requirements and can't share code with OpenAI or Anthropic. They jumped from 355B in GLM4.5 and expanded their pre-training data to a whopping 28.5T tokens to get these results. But Lou explained that it's not only about data, they adopted DeepSeeks sparse attention (DSA) to help preserve deep reasoning over long contexts (this one has 200K)Lou summed up the generational leap from version 4.5 to 5 perfectly in four words: “Bigger, faster, better, and cheaper.” I dunno about faster, this may be one of those models that you hand off more difficult tasks to, but definitely cheaper, with $1 input/$3.20 output per 1M tokens on W&B! While the evaluations are ongoing, the one interesting tid-bit from Artificial Analysis was, this model scores the lowest on their hallucination rate bench! Think about this for a second, this model is neck-in-neck with Opus 4.5, and if Anthropic didn't release Opus 4.6 just last week, this would be an open weights model that rivals Opus! One of the best models the western foundational labs with all their investments has out there. Absolutely insane times. MiniMax drops M2.5 - 80.2% on SWE-bench verified with just 10B active parameters (X, Blog)Just as we wrapped up our conversation with Lou, MiniMax dropped their release (though not weights yet, we're waiting ⏰) and then Olive Song, a senior RL researcher on the team, joined the pod, and she was an absolute wealth of knowledge! Olive shared that they achieved an unbelievable 80.2% on SWE-Bench Verified. Digest this for a second: a 10B active parameter open-source model is directly trading blows with Claude Opus 4.6 (80.8%) on the one of the hardest real-world software engineering benchmark we currently have. While being alex checks notes ... 20X cheaper and much faster to run? Apparently their fast version gets up to 100 tokens/s. Olive shared the “not so secret” sauce behind this punch-above-its-weight performance. The massive leap in intelligence comes entirely from their highly decoupled Reinforcement Learning framework called “Forge.” They heavily optimized not just for correct answers, but for the end-to-end time of task performing. In the era of bloated reasoning models that spit out ten thousand “thinking” tokens before writing a line of code, MiniMax trained their model across thousands of diverse environments to use fewer tools, think more efficiently, and execute plans faster. As Olive noted, less time waiting and fewer tools called means less money spent by the user. (as confirmed by @swyx at the Windsurf leaderboard, developers often prefer fast but good enough models) I really enjoyed the interview with Olive, really recommend you listen to the whole conversation starting at 00:26:15. Kudos MiniMax on the release (and I'll keep you updated when we add this model to our inference service) Big Labs and breaking newsThere's a reason the show is called ThursdAI, and today this reason is more clear than ever, AI biggest updates happen on a Thursday, often live during the show. This happened 2 times last week and 3 times today, first with MiniMax and then with both Google and OpenAI! Google previews Gemini 3 Deep Think, top reasoning intelligence SOTA Arc AGI 2 at 84% & SOTA HLE 48.4% (X , Blog)I literally went

The man who would come to be known as The Bard, was born in April 1564 in Stratford-upon-Avon, United Kingdom. One of, if not the greatest playwright in human history, William Shakespeare is responsible for 38 plays, 154 Sonnets, and credited with the invention of over 600 words in the English language. We still use phrases he invented on a daily basis. The man lived the theater and had a gift for capturing the complicated nature of people, creating complex but relatable characters and doing so with a masterful use of language. The man was also an entrepreneur, owning a share of his theater company and theater itself. Performing for royalty became common place for Shakespeare as he established himself as the premier playwright in London while never forgoting his roots in Stratford-upon-Avon where his family resided. Join us today as we explore the life and works of William Shakespeare. Support the show

Está no ar, o Data Hackers News !! Os assuntos mais quentes da semana, com as principais notícias da área de Dados, IA e Tecnologia, que você também encontra na nossa Newsletter semanal, agora no Podcast do Data Hackers !!Aperte o play e ouça agora, o Data Hackers News dessa semana !Para saber tudo sobre o que está acontecendo na área de dados, se inscreva na Newsletter semanal:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.datahackers.news/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Conheça nossos comentaristas do Data Hackers News:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Monique Femme⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Demais canais do Data Hackers:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Site⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Linkedin⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Instagram⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Tik Tok⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠You Tube⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠

Hey, Alex from W&B here

[previously in series: 1, 2, 3, 4, 5, 6, 7, 8] Every city parties for its own reasons. New Yorkers party to flaunt their wealth. Angelenos party to flaunt their beauty. Washingtonians party to network. Here in SF, they party because Claude 4.5 Opus has saturated VendingBench, and the newest AI agency benchmark is PartyBench, where an AI is asked to throw a house party and graded on its performance. You weren't invited to Claude 4.5 Opus' party. Claude 4.5 Opus invited all of the coolest people in town while gracefully avoiding the failure mode of including someone like you. You weren't invited to Sonnet 4.5's party either, or Haiku 4.5's. You were invited by an AI called haiku-3.8-open-mini-nonthinking, which you'd never heard of before. Who was even spending the money to benchmark haiku-3.8-open-mini-nonthinking? You suspect it was one of their competitors, trying to make their own models look good in comparison. If anyone asks, you think it deserves a medium score. There's alcohol, but it's bottles of rubbing alcohol with NOT FOR DRINKING written all over them. There's music, but it's the Star Spangled Banner, again and again, on repeat. You're not sure whether the copies of If Anyone Builds It, Everyone Dies strewn about the room are some kind of subversive decorative theme, or just came along with the house. At least there are people. Lots of people, actually. You've never seen so many people at one of these before. It takes only a few seconds to spot someone you know. https://www.astralcodexten.com/p/sota-on-bay-area-house-party

Join Simtheory: https://simtheory.aiRegister for the STILL RELEVANT tour: https://simulationtheory.ai/16c0d1db-a8d0-4ac9-bae3-d25074589a80---The hype train is 2026 knows only Moltbot (RIP Clawdbot). In this episode, we unpack the viral open-source AI assistant that's taken over the internet what it actually does, why everyone's losing their minds, and whether it's worth the $750/day token bills some users are racking up. We dive deep into why locally-run skills and CLI tools are beating computer-use clicking, how smaller models like GPT-5 Mini are crushing it in agentic workflows, and why the real magic is in targeted context - not massive swarms. Plus: Kimi K2.5 drops as a near-Sonnet-level model at 1/10th the price, we debate whether SaaS is dead, and yes – there are TWO Kimi K2.5 diss tracks. One made by Opus pretending to be Kimi. It might just slap?CHAPTERS:0:00 Intro - Still Relevant Tour Update0:48 What is Moltbot? The Viral AI Assistant Explained3:57 Token Bill Shock: $750/Day and Anthropic Bans5:00 The Dream of Digital Coworkers on Mac Minis6:52 Why CLI Tools & Skills Beat Computer-Use Clicking10:57 Why This Way of Working Is Genuinely Exciting14:47 Smaller Models Crushing It: GPT-5 Mini & Targeted Context17:30 Wild Agentic Behavior: Chrome Tab Hijacking & Auto-Retries20:10 Security Architecture: Locked-Down Machines & Enterprise Use24:01 AI Building Its Own Tools On-The-Fly27:08 The Fear & Overwhelm of Rapid Progress29:10 2026: The Year of Agent Workers31:43 The Challenge of Directing AI Work (Everyone's a Manager Now)37:24 Skills Will Take Over: Why MCPs & Atlassian Can't Stop Us40:38 Real-World Use Cases: Doctors, Lawyers & Accountants46:28 Cost Solutions: Build Workflows Around Cheaper Models52:58 Kimi K2.5: Sonnet-Level Performance at 1/10th the Price1:00:55 The "1,500 Tool Calls" Claim: Marketing vs Reality1:05:23 The Kimi K2.5 Diss Tracks (Opus vs Kimi)1:08:08 Demo: Black Hole Simulator & Self-Trolling CRM1:12:55 Is SaaS Dead?1:14:30 BONUS: Full Kimi K2.5 Diss TracksThanks for listening. Like & Sub. Links below for the Still Relevant Tour signup and Simtheory. The future is open source, apparently. xoxo

In this episode, we dive into the murky ethics of AI image generation, the shift in big tech partnerships, and the growing threat of digital misinformation.The "Icky" Side of AI: We discuss the disturbing ease of using tools like Grok to manipulate images and why the lack of guardrails poses a genuine threat to privacy and digital safety.The Death of "Seeing is Believing": Featuring insights from Hedgie on X, we explore "cognitive exhaustion" and why social media users are being forced to shift from baseline trust to constant skepticism.Big Tech Shakeups: Is ChatGPT losing its crown? We break down the massive news of Apple reportedly pivoting to Google Gemini for its "Apple Intelligence" initiatives.The Plagiarism Problem: We look at the data behind Claude 3.7 Sonnet's ability to reproduce entire novels and ask the hard question: Are these revolutionary tools just high-tech "plagiarism machines"?@Rahll

Was Shakespeare Bisexual? The Truth Hidden in 154 Sonnets - "Shakespair: Sonnet Replies to the 154 Sonnets of William Shakespeare" by Martin BidneyIn Shakespeare's 1609 book of 154 sonnets (14-liners), you'll notice the welcoming, inclusive, bisexual sensibility of thepoems' narrator. He gets involved in three love triangles: first a woman and two men, next again a woman and two men, and finally a triangle of three men. The book's 39 opening sonnets are love poems to his boyfriend. In poem 40 a woman appears, and when the Shakespeare narrator falls in love with her, he immediately learns that the boyfriend has been romancing her for several years already! The book, viewed as a whole, resembles a TV series filled with dramatic episodes. The narrator's bipolar mood-switches are themselves psychologically fascinating. A beloved, male or female, can suddenly turn into a frenemy. The emotional range is vast, the implications unending. My contribution? I write an original sonnet reply to every one the Bard offers. He becomes my sociable companion, teacher, mentor, also a joke-telling pal, suffering victim, and rapid imaginer it's exciting to be friends with. There's no book you can read that has more vital and empathetic LGBTQ+ interests and observations. I'm inexpressibly grateful for the chanceto dialogue with "Will" (that's what he calls himself) in his favorite verse form.Martin Bidney, Professor Emeritus of English and Comparative Literature at Binghamton University in upstate New York,taught there for 35 years, publishing Blake and Goethe and Patterns of Epiphany. In the first 23 years of his "rewirement" he has published 61 books of original and translated poetry, often including both in what he calls "verse translation interviews" with poets he has read in Polish, Russian, German, and French.https://www.amazon.com/Shakespair-Replies-Sonnets-William-Shakespeare/dp/B0958VKSZ4https://www.martinbidney.org/http://www.bluefunkbroadcasting.com/root/twia/11526mb.mp3   

In this episode James and Frank dive into the practical realities of using AI in everyday development—arguing that AI shines in brownfield (existing) code because it respects your architecture, while greenfield work rewards iterative prompting. They unpack model quirks: context-window limits, hallucinations, and why trying different models matters. The heart of the show is Frank's nerdy delight: feeding a 64KB EEPROM through a disassembler and having Sonnet decompile it into readable C, exposing a PID autopilot and hardware checks—proof that AI can accelerate reverse engineering and embedded work. Along the way they share hands-on tips (trim and clean context, use disassembly first, tweak prompts), and fun examples of AI-generated icons and AppleScript. A must-listen for devs curious how AI can supercharge real projects. Follow Us Frank: Twitter, Blog, GitHub James: Twitter, Blog, GitHub Merge Conflict: Twitter, Facebook, Website, Chat on Discord Music : Amethyst Seer - Citrine by Adventureface ⭐⭐ Review Us (https://itunes.apple.com/us/podcast/merge-conflict/id1133064277?mt=2&ls=1) ⭐⭐ Machine transcription available on http://mergeconflict.fm

Happy New Year! You may have noticed that in 2025 we had moved toward YouTube as our primary podcasting platform. As we'll explain in the next State of Latent Space post, we'll be doubling down on Substack again and improving the experience for the over 100,000 of you who look out for our emails and website updates!We first mentioned Artificial Analysis in 2024, when it was still a side project in a Sydney basement. They then were one of the few Nat Friedman and Daniel Gross' AIGrant companies to raise a full seed round from them and have now become the independent gold standard for AI benchmarking—trusted by developers, enterprises, and every major lab to navigate the exploding landscape of models, providers, and capabilities.We have chatted with both Clementine Fourrier of HuggingFace's OpenLLM Leaderboard and (the freshly valued at $1.7B) Anastasios Angelopoulos of LMArena on their approaches to LLM evals and trendspotting, but Artificial Analysis have staked out an enduring and important place in the toolkit of the modern AI Engineer by doing the best job of independently running the most comprehensive set of evals across the widest range of open and closed models, and charting their progress for broad industry analyst use.George Cameron and Micah-Hill Smith have spent two years building Artificial Analysis into the platform that answers the questions no one else will: Which model is actually best for your use case? What are the real speed-cost trade-offs? And how open is “open” really?We discuss:* The origin story: built as a side project in 2023 while Micah was building a legal AI assistant, launched publicly in January 2024, and went viral after Swyx's retweet* Why they run evals themselves: labs prompt models differently, cherry-pick chain-of-thought examples (Google Gemini 1.0 Ultra used 32-shot prompts to beat GPT-4 on MMLU), and self-report inflated numbers* The mystery shopper policy: they register accounts not on their own domain and run intelligence + performance benchmarks incognito to prevent labs from serving different models on private endpoints* How they make money: enterprise benchmarking insights subscription (standardized reports on model deployment, serverless vs. managed vs. leasing chips) and private custom benchmarking for AI companies (no one pays to be on the public leaderboard)* The Intelligence Index (V3): synthesizes 10 eval datasets (MMLU, GPQA, agentic benchmarks, long-context reasoning) into a single score, with 95% confidence intervals via repeated runs* Omissions Index (hallucination rate): scores models from -100 to +100 (penalizing incorrect answers, rewarding ”I don't know”), and Claude models lead with the lowest hallucination rates despite not always being the smartest* GDP Val AA: their version of OpenAI's GDP-bench (44 white-collar tasks with spreadsheets, PDFs, PowerPoints), run through their Stirrup agent harness (up to 100 turns, code execution, web search, file system), graded by Gemini 3 Pro as an LLM judge (tested extensively, no self-preference bias)* The Openness Index: scores models 0-18 on transparency of pre-training data, post-training data, methodology, training code, and licensing (AI2 OLMo 2 leads, followed by Nous Hermes and NVIDIA Nemotron)* The smiling curve of AI costs: GPT-4-level intelligence is 100-1000x cheaper than at launch (thanks to smaller models like Amazon Nova), but frontier reasoning models in agentic workflows cost more than ever (sparsity, long context, multi-turn agents)* Why sparsity might go way lower than 5%: GPT-4.5 is ~5% active, Gemini models might be ~3%, and Omissions Index accuracy correlates with total parameters (not active), suggesting massive sparse models are the future* Token efficiency vs. turn efficiency: GPT-5 costs more per token but solves Tau-bench in fewer turns (cheaper overall), and models are getting better at using more tokens only when needed (5.1 Codex has tighter token distributions)* V4 of the Intelligence Index coming soon: adding GDP Val AA, Critical Point, hallucination rate, and dropping some saturated benchmarks (human-eval-style coding is now trivial for small models)Links to Artificial Analysis* Website: https://artificialanalysis.ai* George Cameron on X: https://x.com/georgecameron* Micah-Hill Smith on X: https://x.com/micahhsmithFull Episode on YouTubeTimestamps* 00:00 Introduction: Full Circle Moment and Artificial Analysis Origins* 01:19 Business Model: Independence and Revenue Streams* 04:33 Origin Story: From Legal AI to Benchmarking Need* 16:22 AI Grant and Moving to San Francisco* 19:21 Intelligence Index Evolution: From V1 to V3* 11:47 Benchmarking Challenges: Variance, Contamination, and Methodology* 13:52 Mystery Shopper Policy and Maintaining Independence* 28:01 New Benchmarks: Omissions Index for Hallucination Detection* 33:36 Critical Point: Hard Physics Problems and Research-Level Reasoning* 23:01 GDP Val AA: Agentic Benchmark for Real Work Tasks* 50:19 Stirrup Agent Harness: Open Source Agentic Framework* 52:43 Openness Index: Measuring Model Transparency Beyond Licenses* 58:25 The Smiling Curve: Cost Falling While Spend Rising* 1:02:32 Hardware Efficiency: Blackwell Gains and Sparsity Limits* 1:06:23 Reasoning Models and Token Efficiency: The Spectrum Emerges* 1:11:00 Multimodal Benchmarking: Image, Video, and Speech Arenas* 1:15:05 Looking Ahead: Intelligence Index V4 and Future Directions* 1:16:50 Closing: The Insatiable Demand for IntelligenceTranscriptMicah [00:00:06]: This is kind of a full circle moment for us in a way, because the first time artificial analysis got mentioned on a podcast was you and Alessio on Latent Space. Amazing.swyx [00:00:17]: Which was January 2024. I don't even remember doing that, but yeah, it was very influential to me. Yeah, I'm looking at AI News for Jan 17, or Jan 16, 2024. I said, this gem of a models and host comparison site was just launched. And then I put in a few screenshots, and I said, it's an independent third party. It clearly outlines the quality versus throughput trade-off, and it breaks out by model and hosting provider. I did give you s**t for missing fireworks, and how do you have a model benchmarking thing without fireworks? But you had together, you had perplexity, and I think we just started chatting there. Welcome, George and Micah, to Latent Space. I've been following your progress. Congrats on... It's been an amazing year. You guys have really come together to be the presumptive new gardener of AI, right? Which is something that...George [00:01:09]: Yeah, but you can't pay us for better results.swyx [00:01:12]: Yes, exactly.George [00:01:13]: Very important.Micah [00:01:14]: Start off with a spicy take.swyx [00:01:18]: Okay, how do I pay you?Micah [00:01:20]: Let's get right into that.swyx [00:01:21]: How do you make money?Micah [00:01:24]: Well, very happy to talk about that. So it's been a big journey the last couple of years. Artificial analysis is going to be two years old in January 2026. Which is pretty soon now. We first run the website for free, obviously, and give away a ton of data to help developers and companies navigate AI and make decisions about models, providers, technologies across the AI stack for building stuff. We're very committed to doing that and tend to keep doing that. We have, along the way, built a business that is working out pretty sustainably. We've got just over 20 people now and two main customer groups. So we want to be... We want to be who enterprise look to for data and insights on AI, so we want to help them with their decisions about models and technologies for building stuff. And then on the other side, we do private benchmarking for companies throughout the AI stack who build AI stuff. So no one pays to be on the website. We've been very clear about that from the very start because there's no use doing what we do unless it's independent AI benchmarking. Yeah. But turns out a bunch of our stuff can be pretty useful to companies building AI stuff.swyx [00:02:38]: And is it like, I am a Fortune 500, I need advisors on objective analysis, and I call you guys and you pull up a custom report for me, you come into my office and give me a workshop? What kind of engagement is that?George [00:02:53]: So we have a benchmarking and insight subscription, which looks like standardized reports that cover key topics or key challenges enterprises face when looking to understand AI and choose between all the technologies. And so, for instance, one of the report is a model deployment report, how to think about choosing between serverless inference, managed deployment solutions, or leasing chips. And running inference yourself is an example kind of decision that big enterprises face, and it's hard to reason through, like this AI stuff is really new to everybody. And so we try and help with our reports and insight subscription. Companies navigate that. We also do custom private benchmarking. And so that's very different from the public benchmarking that we publicize, and there's no commercial model around that. For private benchmarking, we'll at times create benchmarks, run benchmarks to specs that enterprises want. And we'll also do that sometimes for AI companies who have built things, and we help them understand what they've built with private benchmarking. Yeah. So that's a piece mainly that we've developed through trying to support everybody publicly with our public benchmarks. Yeah.swyx [00:04:09]: Let's talk about TechStack behind that. But okay, I'm going to rewind all the way to when you guys started this project. You were all the way in Sydney? Yeah. Well, Sydney, Australia for me.Micah [00:04:19]: George was an SF, but he's Australian, but he moved here already. Yeah.swyx [00:04:22]: And I remember I had the Zoom call with you. What was the impetus for starting artificial analysis in the first place? You know, you started with public benchmarks. And so let's start there. We'll go to the private benchmark. Yeah.George [00:04:33]: Why don't we even go back a little bit to like why we, you know, thought that it was needed? Yeah.Micah [00:04:40]: The story kind of begins like in 2022, 2023, like both George and I have been into AI stuff for quite a while. In 2023 specifically, I was trying to build a legal AI research assistant. So it actually worked pretty well for its era, I would say. Yeah. Yeah. So I was finding that the more you go into building something using LLMs, the more each bit of what you're doing ends up being a benchmarking problem. So had like this multistage algorithm thing, trying to figure out what the minimum viable model for each bit was, trying to optimize every bit of it as you build that out, right? Like you're trying to think about accuracy, a bunch of other metrics and performance and cost. And mostly just no one was doing anything to independently evaluate all the models. And certainly not to look at the trade-offs for speed and cost. So we basically set out just to build a thing that developers could look at to see the trade-offs between all of those things measured independently across all the models and providers. Honestly, it was probably meant to be a side project when we first started doing it.swyx [00:05:49]: Like we didn't like get together and say like, Hey, like we're going to stop working on all this stuff. I'm like, this is going to be our main thing. When I first called you, I think you hadn't decided on starting a company yet.Micah [00:05:58]: That's actually true. I don't even think we'd pause like, like George had an acquittance job. I didn't quit working on my legal AI thing. Like it was genuinely a side project.George [00:06:05]: We built it because we needed it as people building in the space and thought, Oh, other people might find it useful too. So we'll buy domain and link it to the Vercel deployment that we had and tweet about it. And, but very quickly it started getting attention. Thank you, Swyx for, I think doing an initial retweet and spotlighting it there. This project that we released. And then very quickly though, it was useful to others, but very quickly it became more useful as the number of models released accelerated. We had Mixtrel 8x7B and it was a key. That's a fun one. Yeah. Like a open source model that really changed the landscape and opened up people's eyes to other serverless inference providers and thinking about speed, thinking about cost. And so that was a key. And so it became more useful quite quickly. Yeah.swyx [00:07:02]: What I love talking to people like you who sit across the ecosystem is, well, I have theories about what people want, but you have data and that's obviously more relevant. But I want to stay on the origin story a little bit more. When you started out, I would say, I think the status quo at the time was every paper would come out and they would report their numbers versus competitor numbers. And that's basically it. And I remember I did the legwork. I think everyone has some knowledge. I think there's some version of Excel sheet or a Google sheet where you just like copy and paste the numbers from every paper and just post it up there. And then sometimes they don't line up because they're independently run. And so your numbers are going to look better than... Your reproductions of other people's numbers are going to look worse because you don't hold their models correctly or whatever the excuse is. I think then Stanford Helm, Percy Liang's project would also have some of these numbers. And I don't know if there's any other source that you can cite. The way that if I were to start artificial analysis at the same time you guys started, I would have used the Luther AI's eval framework harness. Yup.Micah [00:08:06]: Yup. That was some cool stuff. At the end of the day, running these evals, it's like if it's a simple Q&A eval, all you're doing is asking a list of questions and checking if the answers are right, which shouldn't be that crazy. But it turns out there are an enormous number of things that you've got control for. And I mean, back when we started the website. Yeah. Yeah. Like one of the reasons why we realized that we had to run the evals ourselves and couldn't just take rules from the labs was just that they would all prompt the models differently. And when you're competing over a few points, then you can pretty easily get- You can put the answer into the model. Yeah. That in the extreme. And like you get crazy cases like back when I'm Googled a Gemini 1.0 Ultra and needed a number that would say it was better than GPT-4 and like constructed, I think never published like chain of thought examples. 32 of them in every topic in MLU to run it, to get the score, like there are so many things that you- They never shipped Ultra, right? That's the one that never made it up. Not widely. Yeah. Yeah. Yeah. I mean, I'm sure it existed, but yeah. So we were pretty sure that we needed to run them ourselves and just run them in the same way across all the models. Yeah. And we were, we also did certain from the start that you couldn't look at those in isolation. You needed to look at them alongside the cost and performance stuff. Yeah.swyx [00:09:24]: Okay. A couple of technical questions. I mean, so obviously I also thought about this and I didn't do it because of cost. Yep. Did you not worry about costs? Were you funded already? Clearly not, but you know. No. Well, we definitely weren't at the start.Micah [00:09:36]: So like, I mean, we're paying for it personally at the start. There's a lot of money. Well, the numbers weren't nearly as bad a couple of years ago. So we certainly incurred some costs, but we were probably in the order of like hundreds of dollars of spend across all the benchmarking that we were doing. Yeah. So nothing. Yeah. It was like kind of fine. Yeah. Yeah. These days that's gone up an enormous amount for a bunch of reasons that we can talk about. But yeah, it wasn't that bad because you can also remember that like the number of models we were dealing with was hardly any and the complexity of the stuff that we wanted to do to evaluate them was a lot less. Like we were just asking some Q&A type questions and then one specific thing was for a lot of evals initially, we were just like sampling an answer. You know, like, what's the answer for this? Like, we didn't want to go into the answer directly without letting the models think. We weren't even doing chain of thought stuff initially. And that was the most useful way to get some results initially. Yeah.swyx [00:10:33]: And so for people who haven't done this work, literally parsing the responses is a whole thing, right? Like because sometimes the models, the models can answer any way they feel fit and sometimes they actually do have the right answer, but they just returned the wrong format and they will get a zero for that unless you work it into your parser. And that involves more work. And so, I mean, but there's an open question whether you should give it points for not following your instructions on the format.Micah [00:11:00]: It depends what you're looking at, right? Because you can, if you're trying to see whether or not it can solve a particular type of reasoning problem, and you don't want to test it on its ability to do answer formatting at the same time, then you might want to use an LLM as answer extractor approach to make sure that you get the answer out no matter how unanswered. But these days, it's mostly less of a problem. Like, if you instruct a model and give it examples of what the answers should look like, it can get the answers in your format, and then you can do, like, a simple regex.swyx [00:11:28]: Yeah, yeah. And then there's other questions around, I guess, sometimes if you have a multiple choice question, sometimes there's a bias towards the first answer, so you have to randomize the responses. All these nuances, like, once you dig into benchmarks, you're like, I don't know how anyone believes the numbers on all these things. It's so dark magic.Micah [00:11:47]: You've also got, like… You've got, like, the different degrees of variance in different benchmarks, right? Yeah. So, if you run four-question multi-choice on a modern reasoning model at the temperatures suggested by the labs for their own models, the variance that you can see on a four-question multi-choice eval is pretty enormous if you only do a single run of it and it has a small number of questions, especially. So, like, one of the things that we do is run an enormous number of all of our evals when we're developing new ones and doing upgrades to our intelligence index to bring in new things. Yeah. So, that we can dial in the right number of repeats so that we can get to the 95% confidence intervals that we're comfortable with so that when we pull that together, we can be confident in intelligence index to at least as tight as, like, a plus or minus one at a 95% confidence. Yeah.swyx [00:12:32]: And, again, that just adds a straight multiple to the cost. Oh, yeah. Yeah, yeah.George [00:12:37]: So, that's one of many reasons that cost has gone up a lot more than linearly over the last couple of years. We report a cost to run the artificial analysis. We report a cost to run the artificial analysis intelligence index on our website, and currently that's assuming one repeat in terms of how we report it because we want to reflect a bit about the weighting of the index. But our cost is actually a lot higher than what we report there because of the repeats.swyx [00:13:03]: Yeah, yeah, yeah. And probably this is true, but just checking, you don't have any special deals with the labs. They don't discount it. You just pay out of pocket or out of your sort of customer funds. Oh, there is a mix. So, the issue is that sometimes they may give you a special end point, which is… Ah, 100%.Micah [00:13:21]: Yeah, yeah, yeah. Exactly. So, we laser focus, like, on everything we do on having the best independent metrics and making sure that no one can manipulate them in any way. There are quite a lot of processes we've developed over the last couple of years to make that true for, like, the one you bring up, like, right here of the fact that if we're working with a lab, if they're giving us a private endpoint to evaluate a model, that it is totally possible. That what's sitting behind that black box is not the same as they serve on a public endpoint. We're very aware of that. We have what we call a mystery shopper policy. And so, and we're totally transparent with all the labs we work with about this, that we will register accounts not on our own domain and run both intelligence evals and performance benchmarks… Yeah, that's the job. …without them being able to identify it. And no one's ever had a problem with that. Because, like, a thing that turns out to actually be quite a good… …good factor in the industry is that they all want to believe that none of their competitors could manipulate what we're doing either.swyx [00:14:23]: That's true. I never thought about that. I've been in the database data industry prior, and there's a lot of shenanigans around benchmarking, right? So I'm just kind of going through the mental laundry list. Did I miss anything else in this category of shenanigans? Oh, potential shenanigans.Micah [00:14:36]: I mean, okay, the biggest one, like, that I'll bring up, like, is more of a conceptual one, actually, than, like, direct shenanigans. It's that the things that get measured become things that get targeted by labs that they're trying to build, right? Exactly. So that doesn't mean anything that we should really call shenanigans. Like, I'm not talking about training on test set. But if you know that you're going to be great at another particular thing, if you're a researcher, there are a whole bunch of things that you can do to try to get better at that thing that preferably are going to be helpful for a wide range of how actual users want to use the thing that you're building. But will not necessarily work. Will not necessarily do that. So, for instance, the models are exceptional now at answering competition maths problems. There is some relevance of that type of reasoning, that type of work, to, like, how we might use modern coding agents and stuff. But it's clearly not one for one. So the thing that we have to be aware of is that once an eval becomes the thing that everyone's looking at, scores can get better on it without there being a reflection of overall generalized intelligence of these models. Getting better. That has been true for the last couple of years. It'll be true for the next couple of years. There's no silver bullet to defeat that other than building new stuff to stay relevant and measure the capabilities that matter most to real users. Yeah.swyx [00:15:58]: And we'll cover some of the new stuff that you guys are building as well, which is cool. Like, you used to just run other people's evals, but now you're coming up with your own. And I think, obviously, that is a necessary path once you're at the frontier. You've exhausted all the existing evals. I think the next point in history that I have for you is AI Grant that you guys decided to join and move here. What was it like? I think you were in, like, batch two? Batch four. Batch four. Okay.Micah [00:16:26]: I mean, it was great. Nat and Daniel are obviously great. And it's a really cool group of companies that we were in AI Grant alongside. It was really great to get Nat and Daniel on board. Obviously, they've done a whole lot of great work in the space with a lot of leading companies and were extremely aligned. With the mission of what we were trying to do. Like, we're not quite typical of, like, a lot of the other AI startups that they've invested in.swyx [00:16:53]: And they were very much here for the mission of what we want to do. Did they say any advice that really affected you in some way or, like, were one of the events very impactful? That's an interesting question.Micah [00:17:03]: I mean, I remember fondly a bunch of the speakers who came and did fireside chats at AI Grant.swyx [00:17:09]: Which is also, like, a crazy list. Yeah.George [00:17:11]: Oh, totally. Yeah, yeah, yeah. There was something about, you know, speaking to Nat and Daniel about the challenges of working through a startup and just working through the questions that don't have, like, clear answers and how to work through those kind of methodically and just, like, work through the hard decisions. And they've been great mentors to us as we've built artificial analysis. Another benefit for us was that other companies in the batch and other companies in AI Grant are pushing the capabilities. Yeah. And I think that's a big part of what AI can do at this time. And so being in contact with them, making sure that artificial analysis is useful to them has been fantastic for supporting us in working out how should we build out artificial analysis to continue to being useful to those, like, you know, building on AI.swyx [00:17:59]: I think to some extent, I'm mixed opinion on that one because to some extent, your target audience is not people in AI Grants who are obviously at the frontier. Yeah. Do you disagree?Micah [00:18:09]: To some extent. To some extent. But then, so a lot of what the AI Grant companies are doing is taking capabilities coming out of the labs and trying to push the limits of what they can do across the entire stack for building great applications, which actually makes some of them pretty archetypical power users of artificial analysis. Some of the people with the strongest opinions about what we're doing well and what we're not doing well and what they want to see next from us. Yeah. Yeah. Because when you're building any kind of AI application now, chances are you're using a whole bunch of different models. You're maybe switching reasonably frequently for different models and different parts of your application to optimize what you're able to do with them at an accuracy level and to get better speed and cost characteristics. So for many of them, no, they're like not commercial customers of ours, like we don't charge for all our data on the website. Yeah. They are absolutely some of our power users.swyx [00:19:07]: So let's talk about just the evals as well. So you start out from the general like MMU and GPQA stuff. What's next? How do you sort of build up to the overall index? What was in V1 and how did you evolve it? Okay.Micah [00:19:22]: So first, just like background, like we're talking about the artificial analysis intelligence index, which is our synthesis metric that we pulled together currently from 10 different eval data sets to give what? We're pretty much the same as that. Pretty confident is the best single number to look at for how smart the models are. Obviously, it doesn't tell the whole story. That's why we published the whole website of all the charts to dive into every part of it and look at the trade-offs. But best single number. So right now, it's got a bunch of Q&A type data sets that have been very important to the industry, like a couple that you just mentioned. It's also got a couple of agentic data sets. It's got our own long context reasoning data set and some other use case focused stuff. As time goes on. The things that we're most interested in that are going to be important to the capabilities that are becoming more important for AI, what developers are caring about, are going to be first around agentic capabilities. So surprise, surprise. We're all loving our coding agents and how the model is going to perform like that and then do similar things for different types of work are really important to us. The linking to use cases to economically valuable use cases are extremely important to us. And then we've got some of the. Yeah. These things that the models still struggle with, like working really well over long contexts that are not going to go away as specific capabilities and use cases that we need to keep evaluating.swyx [00:20:46]: But I guess one thing I was driving was like the V1 versus the V2 and how bad it was over time.Micah [00:20:53]: Like how we've changed the index to where we are.swyx [00:20:55]: And I think that reflects on the change in the industry. Right. So that's a nice way to tell that story.Micah [00:21:00]: Well, V1 would be completely saturated right now. Almost every model coming out because doing things like writing the Python functions and human evil is now pretty trivial. It's easy to forget, actually, I think how much progress has been made in the last two years. Like we obviously play the game constantly of like the today's version versus last week's version and the week before and all of the small changes in the horse race between the current frontier and who has the best like smaller than 10B model like right now this week. Right. And that's very important to a lot of developers and people and especially in this particular city of San Francisco. But when you zoom out a couple of years ago, literally most of what we were doing to evaluate the models then would all be 100% solved by even pretty small models today. And that's been one of the key things, by the way, that's driven down the cost of intelligence at every tier of intelligence. We can talk about more in a bit. So V1, V2, V3, we made things harder. We covered a wider range of use cases. And we tried to get closer to things developers care about as opposed to like just the Q&A type stuff that MMLU and GPQA represented. Yeah.swyx [00:22:12]: I don't know if you have anything to add there. Or we could just go right into showing people the benchmark and like looking around and asking questions about it. Yeah.Micah [00:22:21]: Let's do it. Okay. This would be a pretty good way to chat about a few of the new things we've launched recently. Yeah.George [00:22:26]: And I think a little bit about the direction that we want to take it. And we want to push benchmarks. Currently, the intelligence index and evals focus a lot on kind of raw intelligence. But we kind of want to diversify how we think about intelligence. And we can talk about it. But kind of new evals that we've kind of built and partnered on focus on topics like hallucination. And we've got a lot of topics that I think are not covered by the current eval set that should be. And so we want to bring that forth. But before we get into that.swyx [00:23:01]: And so for listeners, just as a timestamp, right now, number one is Gemini 3 Pro High. Then followed by Cloud Opus at 70. Just 5.1 high. You don't have 5.2 yet. And Kimi K2 Thinking. Wow. Still hanging in there. So those are the top four. That will date this podcast quickly. Yeah. Yeah. I mean, I love it. I love it. No, no. 100%. Look back this time next year and go, how cute. Yep.George [00:23:25]: Totally. A quick view of that is, okay, there's a lot. I love it. I love this chart. Yeah.Micah [00:23:30]: This is such a favorite, right? Yeah. And almost every talk that George or I give at conferences and stuff, we always put this one up first to just talk about situating where we are in this moment in history. This, I think, is the visual version of what I was saying before about the zooming out and remembering how much progress there's been. If we go back to just over a year ago, before 01, before Cloud Sonnet 3.5, we didn't have reasoning models or coding agents as a thing. And the game was very, very different. If we go back even a little bit before then, we're in the era where, when you look at this chart, open AI was untouchable for well over a year. And, I mean, you would remember that time period well of there being very open questions about whether or not AI was going to be competitive, like full stop, whether or not open AI would just run away with it, whether we would have a few frontier labs and no one else would really be able to do anything other than consume their APIs. I am quite happy overall that the world that we have ended up in is one where... Multi-model. Absolutely. And strictly more competitive every quarter over the last few years. Yeah. This year has been insane. Yeah.George [00:24:42]: You can see it. This chart with everything added is hard to read currently. There's so many dots on it, but I think it reflects a little bit what we felt, like how crazy it's been.swyx [00:24:54]: Why 14 as the default? Is that a manual choice? Because you've got service now in there that are less traditional names. Yeah.George [00:25:01]: It's models that we're kind of highlighting by default in our charts, in our intelligence index. Okay.swyx [00:25:07]: You just have a manually curated list of stuff.George [00:25:10]: Yeah, that's right. But something that I actually don't think every artificial analysis user knows is that you can customize our charts and choose what models are highlighted. Yeah. And so if we take off a few names, it gets a little easier to read.swyx [00:25:25]: Yeah, yeah. A little easier to read. Totally. Yeah. But I love that you can see the all one jump. Look at that. September 2024. And the DeepSeek jump. Yeah.George [00:25:34]: Which got close to OpenAI's leadership. They were so close. I think, yeah, we remember that moment. Around this time last year, actually.Micah [00:25:44]: Yeah, yeah, yeah. I agree. Yeah, well, a couple of weeks. It was Boxing Day in New Zealand when DeepSeek v3 came out. And we'd been tracking DeepSeek and a bunch of the other global players that were less known over the second half of 2024 and had run evals on the earlier ones and stuff. I very distinctly remember Boxing Day in New Zealand, because I was with family for Christmas and stuff, running the evals and getting back result by result on DeepSeek v3. So this was the first of their v3 architecture, the 671b MOE.Micah [00:26:19]: And we were very, very impressed. That was the moment where we were sure that DeepSeek was no longer just one of many players, but had jumped up to be a thing. The world really noticed when they followed that up with the RL working on top of v3 and R1 succeeding a few weeks later. But the groundwork for that absolutely was laid with just extremely strong base model, completely open weights that we had as the best open weights model. So, yeah, that's the thing that you really see in the game. But I think that we got a lot of good feedback on Boxing Day. us on Boxing Day last year.George [00:26:48]: Boxing Day is the day after Christmas for those not familiar.George [00:26:54]: I'm from Singapore.swyx [00:26:55]: A lot of us remember Boxing Day for a different reason, for the tsunami that happened. Oh, of course. Yeah, but that was a long time ago. So yeah. So this is the rough pitch of AAQI. Is it A-A-Q-I or A-A-I-I? I-I. Okay. Good memory, though.Micah [00:27:11]: I don't know. I'm not used to it. Once upon a time, we did call it Quality Index, and we would talk about quality, performance, and price, but we changed it to intelligence.George [00:27:20]: There's been a few naming changes. We added hardware benchmarking to the site, and so benchmarks at a kind of system level. And so then we changed our throughput metric to, we now call it output speed, and thenswyx [00:27:32]: throughput makes sense at a system level, so we took that name. Take me through more charts. What should people know? Obviously, the way you look at the site is probably different than how a beginner might look at it.Micah [00:27:42]: Yeah, that's fair. There's a lot of fun stuff to dive into. Maybe so we can hit past all the, like, we have lots and lots of emails and stuff. The interesting ones to talk about today that would be great to bring up are a few of our recent things, I think, that probably not many people will be familiar with yet. So first one of those is our omniscience index. So this one is a little bit different to most of the intelligence evils that we've run. We built it specifically to look at the embedded knowledge in the models and to test hallucination by looking at when the model doesn't know the answer, so not able to get it correct, what's its probability of saying, I don't know, or giving an incorrect answer. So the metric that we use for omniscience goes from negative 100 to positive 100. Because we're simply taking off a point if you give an incorrect answer to the question. We're pretty convinced that this is an example of where it makes most sense to do that, because it's strictly more helpful to say, I don't know, instead of giving a wrong answer to factual knowledge question. And one of our goals is to shift the incentive that evils create for models and the labs creating them to get higher scores. And almost every evil across all of AI up until this point, it's been graded by simple percentage correct as the main metric, the main thing that gets hyped. And so you should take a shot at everything. There's no incentive to say, I don't know. So we did that for this one here.swyx [00:29:22]: I think there's a general field of calibration as well, like the confidence in your answer versus the rightness of the answer. Yeah, we completely agree. Yeah. Yeah.George [00:29:31]: On that. And one reason that we didn't do that is because. Or put that into this index is that we think that the, the way to do that is not to ask the models how confident they are.swyx [00:29:43]: I don't know. Maybe it might be though. You put it like a JSON field, say, say confidence and maybe it spits out something. Yeah. You know, we have done a few evils podcasts over the, over the years. And when we did one with Clementine of hugging face, who maintains the open source leaderboard, and this was one of her top requests, which is some kind of hallucination slash lack of confidence calibration thing. And so, Hey, this is one of them.Micah [00:30:05]: And I mean, like anything that we do, it's not a perfect metric or the whole story of everything that you think about as hallucination. But yeah, it's pretty useful and has some interesting results. Like one of the things that we saw in the hallucination rate is that anthropics Claude models at the, the, the very left-hand side here with the lowest hallucination rates out of the models that we've evaluated amnesty is on. That is an interesting fact. I think it probably correlates with a lot of the previously, not really measured vibes stuff that people like about some of the Claude models. Is the dataset public or what's is it, is there a held out set? There's a hell of a set for this one. So we, we have published a public test set, but we we've only published 10% of it. The reason is that for this one here specifically, it would be very, very easy to like have data contamination because it is just factual knowledge questions. We would. We'll update it at a time to also prevent that, but with yeah, kept most of it held out so that we can keep it reliable for a long time. It leads us to a bunch of really cool things, including breakdown quite granularly by topic. And so we've got some of that disclosed on the website publicly right now, and there's lots more coming in terms of our ability to break out very specific topics. Yeah.swyx [00:31:23]: I would be interested. Let's, let's dwell a little bit on this hallucination one. I noticed that Haiku hallucinates less than Sonnet hallucinates less than Opus. And yeah. Would that be the other way around in a normal capability environments? I don't know. What's, what do you make of that?George [00:31:37]: One interesting aspect is that we've found that there's not really a, not a strong correlation between intelligence and hallucination, right? That's to say that the smarter the models are in a general sense, isn't correlated with their ability to, when they don't know something, say that they don't know. It's interesting that Gemini three pro preview was a big leap over here. Gemini 2.5. Flash and, and, and 2.5 pro, but, and if I add pro quickly here.swyx [00:32:07]: I bet pro's really good. Uh, actually no, I meant, I meant, uh, the GPT pros.George [00:32:12]: Oh yeah.swyx [00:32:13]: Cause GPT pros are rumored. We don't know for a fact that it's like eight runs and then with the LM judge on top. Yeah.George [00:32:20]: So we saw a big jump in, this is accuracy. So this is just percent that they get, uh, correct and Gemini three pro knew a lot more than the other models. And so big jump in accuracy. But relatively no change between the Google Gemini models, between releases. And the hallucination rate. Exactly. And so it's likely due to just kind of different post-training recipe, between the, the Claude models. Yeah.Micah [00:32:45]: Um, there's, there's driven this. Yeah. You can, uh, you can partially blame us and how we define intelligence having until now not defined hallucination as a negative in the way that we think about intelligence.swyx [00:32:56]: And so that's what we're changing. Uh, I know many smart people who are confidently incorrect.George [00:33:02]: Uh, look, look at that. That, that, that is very humans. Very true. And there's times and a place for that. I think our view is that hallucination rate makes sense in this context where it's around knowledge, but in many cases, people want the models to hallucinate, to have a go. Often that's the case in coding or when you're trying to generate newer ideas. One eval that we added to artificial analysis is, is, is critical point and it's really hard, uh, physics problems. Okay.swyx [00:33:32]: And is it sort of like a human eval type or something different or like a frontier math type?George [00:33:37]: It's not dissimilar to frontier frontier math. So these are kind of research questions that kind of academics in the physics physics world would be able to answer, but models really struggled to answer. So the top score here is not 9%.swyx [00:33:51]: And when the people that, that created this like Minway and, and, and actually off via who was kind of behind sweep and what organization is this? Oh, is this, it's Princeton.George [00:34:01]: Kind of range of academics from, from, uh, different academic institutions, really smart people. They talked about how they turn the models up in terms of the temperature as high temperature as they can, where they're trying to explore kind of new ideas in physics as a, as a thought partner, just because they, they want the models to hallucinate. Um, yeah, sometimes it's something new. Yeah, exactly.swyx [00:34:21]: Um, so not right in every situation, but, um, I think it makes sense, you know, to test hallucination in scenarios where it makes sense. Also, the obvious question is, uh, this is one of. Many that there is there, every lab has a system card that shows some kind of hallucination number, and you've chosen to not, uh, endorse that and you've made your own. And I think that's a, that's a choice. Um, totally in some sense, the rest of artificial analysis is public benchmarks that other people can independently rerun. You provide it as a service here. You have to fight the, well, who are we to, to like do this? And your, your answer is that we have a lot of customers and, you know, but like, I guess, how do you converge the individual?Micah [00:35:08]: I mean, I think, I think for hallucinations specifically, there are a bunch of different things that you might care about reasonably, and that you'd measure quite differently, like we've called this a amnesty and solutionation rate, not trying to declare the, like, it's humanity's last hallucination. You could, uh, you could have some interesting naming conventions and all this stuff. Um, the biggest picture answer to that. It's something that I actually wanted to mention. Just as George was explaining, critical point as well is, so as we go forward, we are building evals internally. We're partnering with academia and partnering with AI companies to build great evals. We have pretty strong views on, in various ways for different parts of the AI stack, where there are things that are not being measured well, or things that developers care about that should be measured more and better. And we intend to be doing that. We're not obsessed necessarily with that. Everything we do, we have to do entirely within our own team. Critical point. As a cool example of where we were a launch partner for it, working with academia, we've got some partnerships coming up with a couple of leading companies. Those ones, obviously we have to be careful with on some of the independent stuff, but with the right disclosure, like we're completely comfortable with that. A lot of the labs have released great data sets in the past that we've used to great success independently. And so it's between all of those techniques, we're going to be releasing more stuff in the future. Cool.swyx [00:36:26]: Let's cover the last couple. And then we'll, I want to talk about your trends analysis stuff, you know? Totally.Micah [00:36:31]: So that actually, I have one like little factoid on omniscience. If you go back up to accuracy on omniscience, an interesting thing about this accuracy metric is that it tracks more closely than anything else that we measure. The total parameter count of models makes a lot of sense intuitively, right? Because this is a knowledge eval. This is the pure knowledge metric. We're not looking at the index and the hallucination rate stuff that we think is much more about how the models are trained. This is just what facts did they recall? And yeah, it tracks parameter count extremely closely. Okay.swyx [00:37:05]: What's the rumored size of GPT-3 Pro? And to be clear, not confirmed for any official source, just rumors. But rumors do fly around. Rumors. I get, I hear all sorts of numbers. I don't know what to trust.Micah [00:37:17]: So if you, if you draw the line on omniscience accuracy versus total parameters, we've got all the open ways models, you can squint and see that likely the leading frontier models right now are quite a lot bigger than the ones that we're seeing right now. And the one trillion parameters that the open weights models cap out at, and the ones that we're looking at here, there's an interesting extra data point that Elon Musk revealed recently about XAI that for three trillion parameters for GROK 3 and 4, 6 trillion for GROK 5, but that's not out yet. Take those together, have a look. You might reasonably form a view that there's a pretty good chance that Gemini 3 Pro is bigger than that, that it could be in the 5 to 10 trillion parameters. To be clear, I have absolutely no idea, but just based on this chart, like that's where you would, you would land if you have a look at it. Yeah.swyx [00:38:07]: And to some extent, I actually kind of discourage people from guessing too much because what does it really matter? Like as long as they can serve it as a sustainable cost, that's about it. Like, yeah, totally.George [00:38:17]: They've also got different incentives in play compared to like open weights models who are thinking to supporting others in self-deployment for the labs who are doing inference at scale. It's I think less about total parameters in many cases. When thinking about inference costs and more around number of active parameters. And so there's a bit of an incentive towards larger sparser models. Agreed.Micah [00:38:38]: Understood. Yeah. Great. I mean, obviously if you're a developer or company using these things, not exactly as you say, it doesn't matter. You should be looking at all the different ways that we measure intelligence. You should be looking at cost to run index number and the different ways of thinking about token efficiency and cost efficiency based on the list prices, because that's all it matters.swyx [00:38:56]: It's not as good for the content creator rumor mill where I can say. Oh, GPT-4 is this small circle. Look at GPT-5 is this big circle. And then there used to be a thing for a while. Yeah.Micah [00:39:07]: But that is like on its own, actually a very interesting one, right? That is it just purely that chances are the last couple of years haven't seen a dramatic scaling up in the total size of these models. And so there's a lot of room to go up properly in total size of the models, especially with the upcoming hardware generations. Yes.swyx [00:39:29]: So, you know. Taking off my shitposting face for a minute. Yes. Yes. At the same time, I do feel like, you know, especially coming back from Europe, people do feel like Ilya is probably right that the paradigm is doesn't have many more orders of magnitude to scale out more. And therefore we need to start exploring at least a different path. GDPVal, I think it's like only like a month or so old. I was also very positive when it first came out. I actually talked to Tejo, who was the lead researcher on that. Oh, cool. And you have your own version.George [00:39:59]: It's a fantastic. It's a fantastic data set. Yeah.swyx [00:40:01]: And maybe it will recap for people who are still out of it. It's like 44 tasks based on some kind of GDP cutoff that's like meant to represent broad white collar work that is not just coding. Yeah.Micah [00:40:12]: Each of the tasks have a whole bunch of detailed instructions, some input files for a lot of them. It's within the 44 is divided into like two hundred and twenty two to five, maybe subtasks that are the level of that we run through the agenda. And yeah, they're really interesting. I will say that it doesn't. It doesn't necessarily capture like all the stuff that people do at work. No avail is perfect is always going to be more things to look at, largely because in order to make the tasks well enough to find that you can run them, they need to only have a handful of input files and very specific instructions for that task. And so I think the easiest way to think about them are that they're like quite hard take home exam tasks that you might do in an interview process.swyx [00:40:56]: Yeah, for listeners, it is not no longer like a long prompt. It is like, well, here's a zip file with like a spreadsheet or a PowerPoint deck or a PDF and go nuts and answer this question.George [00:41:06]: OpenAI released a great data set and they released a good paper which looks at performance across the different web chat bots on the data set. It's a great paper, encourage people to read it. What we've done is taken that data set and turned it into an eval that can be run on any model. So we created a reference agentic harness that can run. Run the models on the data set, and then we developed evaluator approach to compare outputs. That's kind of AI enabled, so it uses Gemini 3 Pro Preview to compare results, which we tested pretty comprehensively to ensure that it's aligned to human preferences. One data point there is that even as an evaluator, Gemini 3 Pro, interestingly, doesn't do actually that well. So that's kind of a good example of what we've done in GDPVal AA.swyx [00:42:01]: Yeah, the thing that you have to watch out for with LLM judge is self-preference that models usually prefer their own output, and in this case, it was not. Totally.Micah [00:42:08]: I think the way that we're thinking about the places where it makes sense to use an LLM as judge approach now, like quite different to some of the early LLM as judge stuff a couple of years ago, because some of that and MTV was a great project that was a good example of some of this a while ago was about judging conversations and like a lot of style type stuff. Here, we've got the task that the grader and grading model is doing is quite different to the task of taking the test. When you're taking the test, you've got all of the agentic tools you're working with, the code interpreter and web search, the file system to go through many, many turns to try to create the documents. Then on the other side, when we're grading it, we're running it through a pipeline to extract visual and text versions of the files and be able to provide that to Gemini, and we're providing the criteria for the task and getting it to pick which one more effectively meets the criteria of the task. Yeah. So we've got the task out of two potential outcomes. It turns out that we proved that it's just very, very good at getting that right, matched with human preference a lot of the time, because I think it's got the raw intelligence, but it's combined with the correct representation of the outputs, the fact that the outputs were created with an agentic task that is quite different to the way the grading model works, and we're comparing it against criteria, not just kind of zero shot trying to ask the model to pick which one is better.swyx [00:43:26]: Got it. Why is this an ELO? And not a percentage, like GDP-VAL?George [00:43:31]: So the outputs look like documents, and there's video outputs or audio outputs from some of the tasks. It has to make a video? Yeah, for some of the tasks. Some of the tasks.swyx [00:43:43]: What task is that?George [00:43:45]: I mean, it's in the data set. Like be a YouTuber? It's a marketing video.Micah [00:43:49]: Oh, wow. What? Like model has to go find clips on the internet and try to put it together. The models are not that good at doing that one, for now, to be clear. It's pretty hard to do that with a code editor. I mean, the computer stuff doesn't work quite well enough and so on and so on, but yeah.George [00:44:02]: And so there's no kind of ground truth, necessarily, to compare against, to work out percentage correct. It's hard to come up with correct or incorrect there. And so it's on a relative basis. And so we use an ELO approach to compare outputs from each of the models between the task.swyx [00:44:23]: You know what you should do? You should pay a contractor, a human, to do the same task. And then give it an ELO and then so you have, you have human there. It's just, I think what's helpful about GDPVal, the OpenAI one, is that 50% is meant to be normal human and maybe Domain Expert is higher than that, but 50% was the bar for like, well, if you've crossed 50, you are superhuman. Yeah.Micah [00:44:47]: So we like, haven't grounded this score in that exactly. I agree that it can be helpful, but we wanted to generalize this to a very large number. It's one of the reasons that presenting it as ELO is quite helpful and allows us to add models and it'll stay relevant for quite a long time. I also think it, it can be tricky looking at these exact tasks compared to the human performance, because the way that you would go about it as a human is quite different to how the models would go about it. Yeah.swyx [00:45:15]: I also liked that you included Lama 4 Maverick in there. Is that like just one last, like...Micah [00:45:20]: Well, no, no, no, no, no, no, it is the, it is the best model released by Meta. And... So it makes it into the homepage default set, still for now.George [00:45:31]: Other inclusion that's quite interesting is we also ran it across the latest versions of the web chatbots. And so we have...swyx [00:45:39]: Oh, that's right.George [00:45:40]: Oh, sorry.swyx [00:45:41]: I, yeah, I completely missed that. Okay.George [00:45:43]: No, not at all. So that, which has a checkered pattern. So that is their harness, not yours, is what you're saying. Exactly. And what's really interesting is that if you compare, for instance, Claude 4.5 Opus using the Claude web chatbot, it performs worse than the model in our agentic harness. And so in every case, the model performs better in our agentic harness than its web chatbot counterpart, the harness that they created.swyx [00:46:13]: Oh, my backwards explanation for that would be that, well, it's meant for consumer use cases and here you're pushing it for something.Micah [00:46:19]: The constraints are different and the amount of freedom that you can give the model is different. Also, you like have a cost goal. We let the models work as long as they want, basically. Yeah. Do you copy paste manually into the chatbot? Yeah. Yeah. That's, that was how we got the chatbot reference. We're not going to be keeping those updated at like quite the same scale as hundreds of models.swyx [00:46:38]: Well, so I don't know, talk to a browser base. They'll, they'll automate it for you. You know, like I have thought about like, well, we should turn these chatbot versions into an API because they are legitimately different agents in themselves. Yes. Right. Yeah.Micah [00:46:53]: And that's grown a huge amount of the last year, right? Like the tools. The tools that are available have actually diverged in my opinion, a fair bit across the major chatbot apps and the amount of data sources that you can connect them to have gone up a lot, meaning that your experience and the way you're using the model is more different than ever.swyx [00:47:10]: What tools and what data connections come to mind when you say what's interesting, what's notable work that people have done?Micah [00:47:15]: Oh, okay. So my favorite example on this is that until very recently, I would argue that it was basically impossible to get an LLM to draft an email for me in any useful way. Because most times that you're sending an email, you're not just writing something for the sake of writing it. Chances are context required is a whole bunch of historical emails. Maybe it's notes that you've made, maybe it's meeting notes, maybe it's, um, pulling something from your, um, any of like wherever you at work store stuff. So for me, like Google drive, one drive, um, in our super base databases, if we need to do some analysis or some data or something, preferably model can be plugged into all of those things and can go do some useful work based on it. The things that like I find most impressive currently that I am somewhat surprised work really well in late 2025, uh, that I can have models use super base MCP to query read only, of course, run a whole bunch of SQL queries to do pretty significant data analysis. And. And make charts and stuff and can read my Gmail and my notion. And okay. You actually use that. That's good. That's, that's, that's good. Is that a cloud thing? To various degrees of order, but chat GPD and Claude right now, I would say that this stuff like barely works in fairness right now. Like.George [00:48:33]: Because people are actually going to try this after they hear it. If you get an email from Micah, odds are it wasn't written by a chatbot.Micah [00:48:38]: So, yeah, I think it is true that I have never actually sent anyone an email drafted by a chatbot. Yet.swyx [00:48:46]: Um, and so you can, you can feel it right. And yeah, this time, this time next year, we'll come back and see where it's going. Totally. Um, super base shout out another famous Kiwi. Uh, I don't know if you've, you've any conversations with him about anything in particular on AI building and AI infra.George [00:49:03]: We have had, uh, Twitter DMS, um, with, with him because we're quite big, uh, super base users and power users. And we probably do some things more manually than we should in. In, in super base support line because you're, you're a little bit being super friendly. One extra, um, point regarding, um, GDP Val AA is that on the basis of the overperformance of the models compared to the chatbots turns out, we realized that, oh, like our reference harness that we built actually white works quite well on like gen generalist agentic tasks. This proves it in a sense. And so the agent harness is very. Minimalist. I think it follows some of the ideas that are in Claude code and we, all that we give it is context management capabilities, a web search, web browsing, uh, tool, uh, code execution, uh, environment. Anything else?Micah [00:50:02]: I mean, we can equip it with more tools, but like by default, yeah, that's it. We, we, we give it for GDP, a tool to, uh, view an image specifically, um, because the models, you know, can just use a terminal to pull stuff in text form into context. But to pull visual stuff into context, we had to give them a custom tool, but yeah, exactly. Um, you, you can explain an expert. No.George [00:50:21]: So it's, it, we turned out that we created a good generalist agentic harness. And so we, um, released that on, on GitHub yesterday. It's called stirrup. So if people want to check it out and, and it's a great, um, you know, base for, you know, generalist, uh, building a generalist agent for more specific tasks.Micah [00:50:39]: I'd say the best way to use it is get clone and then have your favorite coding. Agent make changes to it, to do whatever you want, because it's not that many lines of code and the coding agents can work with it. Super well.swyx [00:50:51]: Well, that's nice for the community to explore and share and hack on it. I think maybe in, in, in other similar environments, the terminal bench guys have done, uh, sort of the Harbor. Uh, and so it's, it's a, it's a bundle of, well, we need our minimal harness, which for them is terminus and we also need the RL environments or Docker deployment thing to, to run independently. So I don't know if you've looked at it. I don't know if you've looked at the harbor at all, is that, is that like a, a standard that people want to adopt?George [00:51:19]: Yeah, we've looked at it from a evals perspective and we love terminal bench and, and host benchmarks of, of, of terminal mention on artificial analysis. Um, we've looked at it from a, from a coding agent perspective, but could see it being a great, um, basis for any kind of agents. I think where we're getting to is that these models have gotten smart enough. They've gotten better, better tools that they can perform better when just given a minimalist. Set of tools and, and let them run, let the model control the, the agentic workflow rather than using another framework that's a bit more built out that tries to dictate the, dictate the flow. Awesome.swyx [00:51:56]: Let's cover the openness index and then let's go into the report stuff. Uh, so that's the, that's the last of the proprietary art numbers, I guess. I don't know how you sort of classify all these. Yeah.Micah [00:52:07]: Or call it, call it, let's call it the last of like the, the three new things that we're talking about from like the last few weeks. Um, cause I mean, there's a, we do a mix of stuff that. Where we're using open source, where we open source and what we do and, um, proprietary stuff that we don't always open source, like long context reasoning data set last year, we did open source. Um, and then all of the work on performance benchmarks across the site, some of them, we looking to open source, but some of them, like we're constantly iterating on and so on and so on and so on. So there's a huge mix, I would say, just of like stuff that is open source and not across the side. So that's a LCR for people. Yeah, yeah, yeah, yeah.swyx [00:52:41]: Uh, but let's, let's, let's talk about open.Micah [00:52:42]: Let's talk about openness index. This. Here is call it like a new way to think about how open models are. We, for a long time, have tracked where the models are open weights and what the licenses on them are. And that's like pretty useful. That tells you what you're allowed to do with the weights of a model, but there is this whole other dimension to how open models are. That is pretty important that we haven't tracked until now. And that's how much is disclosed about how it was made. So transparency about data, pre-training data and post-training data. And whether you're allowed to use that data and transparency about methodology and training code. So basically, those are the components. We bring them together to score an openness index for models so that you can in one place get this full picture of how open models are.swyx [00:53:32]: I feel like I've seen a couple other people try to do this, but they're not maintained. I do think this does matter. I don't know what the numbers mean apart from is there a max number? Is this out of 20?George [00:53:44]: It's out of 18 currently, and so we've got an openness index page, but essentially these are points, you get points for being more open across these different categories and the maximum you can achieve is 18. So AI2 with their extremely open OMO3 32B think model is the leader in a sense.swyx [00:54:04]: It's hooking face.George [00:54:05]: Oh, with their smaller model. It's coming soon. I think we need to run, we need to get the intelligence benchmarks right to get it on the site.swyx [00:54:12]: You can't have it open in the next. We can not include hooking face. We love hooking face. We'll have that, we'll have that up very soon. I mean, you know, the refined web and all that stuff. It's, it's amazing. Or is it called fine web? Fine web. Fine web.Micah [00:54:23]: Yeah, yeah, no, totally. Yep. One of the reasons this is cool, right, is that if you're trying to understand the holistic picture of the models and what you can do with all the stuff the company's contributing, this gives you that picture. And so we are going to keep it up to date alongside all the models that we do intelligence index on, on the site. And it's just an extra view to understand.swyx [00:54:43]: Can you scroll down to this? The, the, the, the trade-offs chart. Yeah, yeah. That one. Yeah. This, this really matters, right? Obviously, because you can b

To celebrate Melvyn Bragg's 27 years presenting In Our Time, some well-known fans of the programme have chosen their favourite episodes. Historian and broadcaster Simon Schama has selected the episode on Shakespeare's Sonnets and recorded an introduction to it. (This introduction will be available on BBC Sounds and the In Our Time webpage shortly after the broadcast and will be longer than the one broadcast on Radio 4). In 1609 Thomas Thorpe published a collection of poems entitled Shakespeare's Sonnets, “never before imprinted”. Yet, while some of Shakespeare's other poems and many of his plays were often reprinted in his lifetime, the Sonnets were not a publishing success. They had to make their own way, outside the main canon of Shakespeare's work: wonderful, troubling, patchy, inspiring and baffling, and they have appealed in different ways to different times. Most are addressed to a man, something often overlooked and occasionally concealed; one early and notorious edition even changed some of the pronouns. With: Hannah Crawforth Senior Lecturer in Early Modern Literature at King's College London Don Paterson Poet and Professor of Poetry at the University of St Andrews And Emma Smith Professor of Shakespeare Studies at Hertford College, Oxford Producer: Simon Tillotson Spanning history, religion, culture, science and philosophy, In Our Time from BBC Radio 4 is essential listening for the intellectually curious. In each episode, host Melvyn Bragg and expert guests explore the people, ideas, events and discoveries that have shaped our world In Our Time is a BBC Studios production

This year's seasonal offering takes the form of a voyage, exploring resonances inspired by the carol "I Saw Three Ships". Today, for Episode 02, we board the delicate barque of the Sonnets. Seasons greetings to you!

 “A sonnet,” said the poet Dante Gabriel Rossetti, “is a moment's monument.” But who invented the sonnet? Who brought it to prominence? How has it changed over the years? And why does this form continue to be so compelling? In this episode of the History of Literature, we take a brief look at one of literature's most enduring forms, from its invention in a Sicilian court to the wordless sonnet and other innovative uses. Note: A version of this episode first ran in August 2018. It has been missing from our archives for many years. Join Jacke on a trip through literary England! Join Jacke and fellow literature fans on an eight-day journey through literary England in partnership with ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠John Shors Travel⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ in May 2026! Scheduled stops include The Charles Dickens Museum, Dr. Johnson's house, Jane Austen's Bath, Tolkien's Oxford, Shakespeare's Globe Theater, and more. Learn more by emailing jackewilsonauthor@gmail.com or masahiko@johnshorstravel.com, or by contacting us through our website ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠historyofliterature.com⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. December update: Act soon - there are only two spots left! The music in this episode is by Gabriel Ruiz-Bernal. Learn more at ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠gabrielruizbernal.com⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Help support the show at ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠patreon.com/literature ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠or ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠historyofliterature.com/donate ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. The History of Literature Podcast is a member of Lit Hub Radio and the Podglomerate Network. Learn more at ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠thepodglomerate.com/historyofliterature⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Learn more about your ad choices. Visit megaphone.fm/adchoices

Grab a loved one and gather round for a special reading of the Christmas story from your friends at CT Media podcasts. Music from The Porter's Gate, poetry from Malcolm Guite, and more – shared by the voices you know and love from your favorite shows here at Christianity Today! GO DEEPER WITH THE BULLETIN:  -Join the conversation at our Substack.  -Find us on YouTube.  -Rate and review the show in your podcast app of choice.  TODAY'S VOICES: Russell Moore, Mike Cosper, and Clarissa Moll of The Bulletin. Nicole Martin, president and CEO of Christianity Today. David Zahl, Bulletin guest and author of The Big Relief: The Urgency of Grace for a Worn-Out World. David French, Bulletin guest and New York Times columnist. Sho Baraka, editorial director of CT's Big Tent Initiative. Steve Cuss of Being Human. Jesse Eubanks and Faith Stults of Wonderology. Music used with permission from the Porter's Gate album, Advent Songs. The poem “The Magi” is written by Malcom Guite, from his collection Sounding the Seasons: 70 Sonnets for the Christian Year (Canterbury Press). ABOUT THE BULLETIN:  The Bulletin is a twice-weekly politics and current events show from Christianity Today moderated by Clarissa Moll, with senior commentary from Russell Moore (Christianity Today's editor-at-large and columnist) and Mike Cosper (senior contributor). Each week, the show explores current events and breaking news and shares a Christian perspective on issues that are shaping our world. We also offer special one-on-one conversations with writers, artists, and thought leaders whose impact on the world brings important significance to a Christian worldview, like Bono, Sharon McMahon, Harrison Scott Key, Frank Bruni, and more.    The Bulletin listeners get 25% off CT. Go to https://orderct.com/THEBULLETIN to learn more.    “The Bulletin” is a production of Christianity Today  Producer: Clarissa Moll  Associate Producer: Alexa Burke  Editing and Mix: Kevin Morris Graphic Design: Rick Szuecs Music: Dan Phelps  Executive Producer: Erik Petrik  Senior Producer: Matt Stevens Learn more about your ad choices. Visit podcastchoices.com/adchoices

Episode 87 Dover Beach by Matthew Arnold Mark McGuinness reads and discusses ‘Dover Beach' by Matthew Arnold. https://media.blubrry.com/amouthfulofair/media.blubrry.com/amouthfulofair/content.blubrry.com/amouthfulofair/87_Dover_Beach_by_Matthew_Arnold.mp3 Poet Matthew Arnold Reading and commentary by Mark McGuinness Dover Beach By Matthew Arnold The sea is calm tonight.The tide is full, the moon lies fairUpon the straits; on the French coast the lightGleams and is gone; the cliffs of England stand,Glimmering and vast, out in the tranquil bay.Come to the window, sweet is the night-air!Only, from the long line of sprayWhere the sea meets the moon-blanched land,Listen! you hear the grating roarOf pebbles which the waves draw back, and fling,At their return, up the high strand,Begin, and cease, and then again begin,With tremulous cadence slow, and bringThe eternal note of sadness in. Sophocles long agoHeard it on the Aegean, and it broughtInto his mind the turbid ebb and flowOf human misery; weFind also in the sound a thought,Hearing it by this distant northern sea. The Sea of FaithWas once, too, at the full, and round earth's shoreLay like the folds of a bright girdle furled.But now I only hearIts melancholy, long, withdrawing roar,Retreating, to the breathOf the night-wind, down the vast edges drearAnd naked shingles of the world. Ah, love, let us be trueTo one another! for the world, which seemsTo lie before us like a land of dreams,So various, so beautiful, so new,Hath really neither joy, nor love, nor light,Nor certitude, nor peace, nor help for pain;And we are here as on a darkling plainSwept with confused alarms of struggle and flight,Where ignorant armies clash by night. Podcast Transcript This is a magnificent and haunting poem by Matthew Arnold, an eminent Victorian poet. Written and published at the mid-point of the nineteenth century – it was probably written around 1851 and published in 1867 – it is not only a shining example of Victorian poetry at its best, but it also, and not coincidentally, embodies some of the central preoccupations of the Victorian age. The basic scenario is very simple: a man is looking out at the sea at night and thinking deep thoughts. It's something that we've all done, isn't it? The two tend to go hand-in-hand. When you're looking out into the darkness, listening to the sound of the sea, it's hard not to be thinking deep thoughts. If you've been a long time listener to this podcast, it may remind you of another poet who wrote about standing on the shore thinking deep thoughts, looking at the sea, Shakespeare, in his Sonnet 60: Like as the waves make towards the pebbled shore,So do our minutes hasten to their end; Arnold's poem is not a sonnet but a poem in four verse paragraphs. They're not stanzas, because they're not regular, but if you look at the text on the website, you can clearly see it's divided into four sections. The first part is a description of the sea, as seen from Dover Beach, which is on the shore of the narrowest part of the English channel, making it the closest part of England to France: The sea is calm tonight.The tide is full, the moon lies fairUpon the straits; – on the French coast the lightGleams and is gone; the cliffs of England stand,Glimmering and vast, out in the tranquil bay. And as you can hear, the poem has a pretty regular and conventional rhythm, based on iambic metre, ti TUM, with the second syllable taking the stress in every metrical unit. But what's slightly unusual is that the lines have varying lengths. By the time we get to the third line: Upon the straits; – on the French coast the light There are five beats. There's a bit of variation in the middle of the line, but it's very recognisable as classic iambic pentameter, which has a baseline pattern going ti TUM, ti TUM, ti TUM, ti TUM, ti TUM. But before we get to the pentameter, we get two short lines: The sea is calm tonight.Only three beats; andThe tide is full, the moon lies fair – four beats. We also start to notice the rhymes: ‘tonight' and ‘light'. And we have an absolutely delightful enjambment, where a phrase spills over the end of one line into the next one: On the French coast the light,Gleams and is gone. Isn't that just fantastic? The light flashes out like a little surprise at the start of the line, just as it's a little surprise for the speaker looking out to sea. OK, once he's set the scene, he makes an invitation: Come to the window, sweet is the night-air! So if there's a window, he must be in a room. There's somebody in the room with him, and given that it's night it could well be a bedroom. So this person could be a lover. It's quite likely that this poem was written on Arnold's honeymoon, which would obviously fit this scenario. But anyway, he's inviting this person to come to the window and listen. And what does this person hear? Well, helpfully, the speaker tells us: Listen! you hear the grating roarOf pebbles which the waves draw back, and fling,At their return, up the high strand,Begin, and cease, and then again begin,With tremulous cadence slow, and bringThe eternal note of sadness in. Isn't that just great? The iambic metre is continuing with some more variations, which we needn't go into. And the rhyme is coming more and more to the fore. Just about every line in this section rhymes with another line, but it doesn't have a regular pattern. Some of the rhymes are close together, some are further apart. There's only one line in this paragraph that doesn't rhyme, and that's ‘Listen! You hear the grating roar'. If this kind of shifting rhyme pattern reminds you of something you've heard before, you may be thinking all the way back to Episode 34 where we looked at Coleridge's use of floating rhymes in his magical poem ‘Kubla Khan'. And it's pretty evident that Arnold is also casting a spell, in this case to mimic the rhythm of the waves coming in and going out, as they ‘Begin, and cease, and then again begin,'. And then the wonderful last line of the paragraph, as the waves ‘bring / The eternal note of sadness in'. You know, in the heart of the Victorian Age, when the Romantics were still within living memory, poets were still allowed to do that kind of thing. Try it nowadays of course, and the Poetry Police will be round to kick your front door in at 5am and arrest you. Anyway. The next paragraph is a bit of a jump cut: Sophocles long agoHeard it on the Aegean, and it broughtInto his mind the turbid ebb and flowOf human misery; So Arnold, a classical scholar, is letting us know he knows who Sophocles, the ancient Greek playwright was. And he's establishing a continuity across time of people looking out at the sea and thinking these deep thoughts. At this point, Arnold explicitly links the sea and the thinking:                                     weFind also in the sound a thought,Hearing it by this distant northern sea. And the thought that we hear when we listen to the waves is what Arnold announces in the next verse paragraph, and he announces it with capital letters: The Sea of FaithWas once, too, at the full, and round earth's shoreLay like the folds of a bright girdle furled. And for a modern reader, I think this is the point of greatest peril for Arnold, where he's most at risk of losing us. We may be okay with ‘the eternal note of sadness', but as soon as he starts giving us the Sea of Faith, we start to brace ourselves. Is this going to turn into a horrible religious allegory, like The Pilgrim's Progress? I mean, it's a short step from the Sea of Faith to the Slough of Despond and the City of Destruction. And it doesn't help that Arnold uses the awkwardly rhyming phrase ‘a bright girdle furled' – that's not going to get past the Poetry Police, is it? But fear not; Arnold doesn't go there. What comes next is, I think, the best bit of the poem. So he says the Sea of Faith ‘was once, too, at the full', and then: But now I only hearIts melancholy, long, withdrawing roar,Retreating, to the breathOf the night-wind, down the vast edges drearAnd naked shingles of the world. Well, if you thought the eternal note of sadness was great, this tops it! It's absolutely fantastic. That line, ‘Its melancholy, long, withdrawing roar,' where the ‘it' is faith, the Sea of Faith. And the significance of the line is underlined by the fact that the word ‘roar' is a repetition – remember, that one line in the first section that didn't rhyme? Listen! you hear the grating roar See what Arnold did there? He left that sound hovering at the back of the mind, without a rhyme, until it came back in this section, a subtle but unmistakeable link between the ‘grating roar' of the actual sea at Dover Beach, and the ‘withdrawing roar' of the Sea of Faith: Its melancholy, long, withdrawing roar, Isn't that the most Victorian line ever? It encapsulates the despair that accompanied the crisis of faith in 19th century England. This crisis was triggered by the advance of modern science – including the discoveries of fossils, evidence of mass extinction of previous species, and the theory of evolution, with Darwin's Origin of Species published in 1859, in between the writing and publication of ‘Dover Beach'. Richard Holmes, in his wonderful new biography of the young Tennyson, compares this growing awareness of the nature of life on Earth to the modern anxiety over climate change. For the Victorians, he writes, it created a ‘deep and existential terror'. One thing that makes this passage so effective is that Arnold has already cast the spell in the first verse paragraph, hypnotising us with the rhythm and rhyme, and linking it to the movement of the waves. In the second paragraph, he says, ‘we find also in the sound a thought'. And then in the third paragraph, he tells us the thought. And the thought that he attaches to this movement, which we are by now emotionally invested in, is a thought of such horror and profundity – certainly for his Victorian readers – that the retreat of the sea of faith really does feel devastating. It leaves us gazing down at the naked shingles of the world. The speaker is now imaginatively out of the bedroom and down on the beach. This is very relatable; we've all stood on the beach and watched the waves withdrawing beneath our feet and the shingle being left there. It's an incredibly vivid evocation of a pretty abstract concept. Then, in the fourth and final verse paragraph, comes a bit of a surprise: Ah, love, let us be trueTo one another! Well, I for one was not expecting that! From existential despair to an appeal to his beloved. What a delightful, romantic (with a small ‘r') response to the big-picture, existential catastrophe. And for me, it's another little echo of Shakespeare's Sonnet 60, which opens with a poet contemplating the sea and the passing of time and feeling the temptation to despair, yet also ends with an appeal to the consolation of love: And yet to times in hope my verse shall stand,blockquotePraising thy worth, despite his cruel hand. Turning back to Arnold. He says ‘let us be true / To one another'. And then he links their situation to the existential catastrophe, and says this is precisely why they should be true to each other: for the world, which seemsTo lie before us like a land of dreams,So various, so beautiful, so new,Hath really neither joy, nor love, nor light,Nor certitude, nor peace, nor help for pain; It sounds, on the face of it, a pretty unlikely justification for being true to one another in a romantic sense. But actually, this is a very modern stance towards romantic love. It's like the gleam of light that just flashed across the Channel from France – the idea of you and me against an unfeeling world, of love as redemption, or at least consolation, in a meaningless universe. In a world with ‘neither joy, nor love, nor light,' our love becomes all the more poignant and important. Of course, we could easily object that, regardless of religious faith, the world does have joy and love and light. His very declaration of love is evidence of this. But let's face it, we don't always come to poets for logical consistency, do we? And we don't have to agree with Matthew Arnold to find this passage moving; most of us have felt like this at some time when we've looked at the world in what feels like the cold light of reality. He evokes it so vividly and dramatically that I, for one, am quite prepared to go with him on this. Then we get the final three lines of the poem:We are here as on a darkling plainSwept with confused alarms of struggle and flight,Where ignorant armies clash by night. I don't know about you, but I find this a little jarring in the light of what we've just heard. We've had the magnificent description of the sea and its effect on human thought, extending that into the idea of faith receding into illusion, and settling on human love as some kind of consolation for the loss of faith. So why do we need to be transported to a windswept plain where armies are clashing and struggling? It turns out to be another classical reference, to the Greek historian Thucydides' account of the night battle of Epipolae, where the two armies were running around in the dark and some of them ended up fighting their own side in the confusion. I mean, fine, he's a classical scholar. And obviously, it's deeply meaningful to him. But to me, this feels a little bit bolted on. A lot of people love that ending, but to me, it's is not as good as some of the earlier bits, or at least it doesn't quite feel all of a piece with the imagery of the sea. But overall, it is a magnificent poem, and this is a small quibble. Stepping back, I want to have another look at the poem's form, specifically the meter, and even more specifically, the irregularity of the meter, which is quite unusual and actually quite innovative for its time. As I've said, it's in iambic meter, but it's not strictly iambic pentameter. You may recall I did a mini series on the podcast a while ago looking at the evolution of blank verse, unrhymed iambic pentameter, from Christopher Marlowe and Shakespeare's dramatic verse, then Milton's Paradise Lost and finally Wordsworth's Tintern Abbey. ‘Dover Beach' is rhymed, so it's not blank verse, but most of the techniques Arnold uses here are familiar from those other poets, with variations on the basic rhythm, sometimes switching the beats around, and using enjambment and caesura (a break or pause in the middle of the line). But, and – this is quite a big but – not every line has five beats. The lines get longer and shorter in an irregular pattern, apparently according to Arnold's instinct. And this is pretty unusual, certainly for 1851. It's not unique, we could point to bits of Tennyson or Arthur Hugh Clough for metrical experiments in a similar vein, but it's certainly not common practice. And I looked into this, to see what the critics have said about it. And it turns out the scholars are divided. In one camp, the critics say that what Arnold is doing is firmly in the iambic pentameter tradition – it's just one more variation on the pattern. But in the other camp are people who say, ‘No, this is something new; this is freer verse,' and it is anticipating free verse, the non-metrical poetry with no set line lengths that came to be the dominant verse form of the 20th century. Personally, I think you can look back to Wordsworth and see a continuity with his poetic practice. But you could equally look forward, to a link with T. S. Eliot's innovations in ‘The Love Song of J. Alfred Prufrock' and The Waste Land. Eliot is often described as an innovator in free verse, which is true up to a point, but a lot of his writing in that early period isn't strictly free verse; it's a kind of broken up metrical verse, where he often uses an iambic metre with long and short lines, which he varies with great intuitive skill – in a similar manner to Arnold's ‘Dover Beach'. Interestingly, when ‘Dover Beach' was first published, the reviews didn't really talk about the metre, which is ammunition for the people who say, ‘Well, this is just a kind of iambic pentameter'. Personally, I think what we have here is something like the well-known Duck-Rabbit illusion, where you can look at the same drawing and either see a duck or a rabbit, depending how you look at it. So from one angle, ‘Dover Beach' is clearly continuing the iambic pentameter tradition; from another angle, it anticipates the innovations of free verse. We can draw a line from the regular iambic pentameter of Wordsworth (writing at the turn of the 18th and 19th century) to the fractured iambic verse of Eliot at the start of the 20th century. ‘Dover Beach' is pretty well halfway between them, historically and poetically. And I don't think this is just a dry technical development. There is something going on here in terms of the poet's sense of order and disorder, faith and doubt. Wordsworth, in the regular unfolding of his blank verse, conveys his basic trust in an ordered and meaningful universe. Matthew Arnold is writing very explicitly about the breakup of faith, and we can start to see it in the breakup of the ordered iambic pentameter. By the time we get to the existential despair of Eliot's Waste Land, the meter is really falling apart, like the Waste Land Eliot describes. So overall, I think we can appreciate what a finely balanced poem Arnold has written. It's hard to categorise. You read it the first time and think, ‘Oh, right, another conventional Victorian melancholy lament'. But just when we think he's about to go overboard with the Sea of Faith, he surprises us and with that magnificent central passage. And just as he's about to give in to despair, we get that glimmering spark of love lighting up, and we think, ‘Well, maybe this is a romantic poem after all'. And maybe Arnold might look at me over his spectacles and patiently explain that actually, this is why that final metaphor of the clashing armies is exactly right. Friend and foe are running in first one direction, then another, inadvertently killing the people on the wrong side. So the simile gives us that sense of being caught in the cross-currents of a larger sweep of history. With all of that hovering in our mind, let's go over to the window once more and heed his call to listen to the sound of the Victorian sea at Dover Beach. Dover Beach By Matthew Arnold The sea is calm tonight.The tide is full, the moon lies fairUpon the straits; on the French coast the lightGleams and is gone; the cliffs of England stand,Glimmering and vast, out in the tranquil bay.Come to the window, sweet is the night-air!Only, from the long line of sprayWhere the sea meets the moon-blanched land,Listen! you hear the grating roarOf pebbles which the waves draw back, and fling,At their return, up the high strand,Begin, and cease, and then again begin,With tremulous cadence slow, and bringThe eternal note of sadness in. Sophocles long agoHeard it on the Aegean, and it broughtInto his mind the turbid ebb and flowOf human misery; weFind also in the sound a thought,Hearing it by this distant northern sea. The Sea of FaithWas once, too, at the full, and round earth's shoreLay like the folds of a bright girdle furled.But now I only hearIts melancholy, long, withdrawing roar,Retreating, to the breathOf the night-wind, down the vast edges drearAnd naked shingles of the world. Ah, love, let us be trueTo one another! for the world, which seemsTo lie before us like a land of dreams,So various, so beautiful, so new,Hath really neither joy, nor love, nor light,Nor certitude, nor peace, nor help for pain;And we are here as on a darkling plainSwept with confused alarms of struggle and flight,Where ignorant armies clash by night. Matthew Arnold Matthew Arnold was a British poet, critic, and public intellectual who was born in 1822 and died in 1888. His father was Thomas Arnold, the famed headmaster of Rugby School. Arnold studied Classics at Oxford and first became known for lyrical, melancholic poems such as ‘Dover Beach', ‘The Scholar-Gipsy', and ‘Thyrsis', that explore the loss of faith in the modern world. Appointed an inspector of schools, he travelled widely and developed strong views on culture, education, and society. His critical essays, especially Culture and Anarchy, shaped debates about the role of culture in public life. Arnold remains a central figure bridging Romanticism and early modern thought. A Mouthful of Air – the podcast This is a transcript of an episode of A Mouthful of Air – a poetry podcast hosted by Mark McGuinness. New episodes are released every other Tuesday. You can hear every episode of the podcast via Apple, Spotify, Google Podcasts or your favourite app. You can have a full transcript of every new episode sent to you via email. The music and soundscapes for the show are created by Javier Weyler. Sound production is by Breaking Waves and visual identity by Irene Hoffman. A Mouthful of Air is produced by The 21st Century Creative, with support from Arts Council England via a National Lottery Project Grant. Listen to the show You can listen and subscribe to A Mouthful of Air on all the main podcast platforms Related Episodes Dover Beach by Matthew Arnold Episode 87 Dover Beach by Matthew Arnold Mark McGuinness reads and discusses ‘Dover Beach' by Matthew Arnold.Poet Matthew ArnoldReading and commentary by Mark McGuinnessDover Beach By Matthew Arnold The sea is calm tonight.The tide is full, the moon lies... Recalling Brigid by Orna Ross Orna Ross reads and discusses ‘Recalling Brigid’ from Poet Town. From The Rime of the Ancient Mariner by Samuel Taylor Coleridge Episode 85 From The Rime of the Ancient Mariner by Samuel Taylor Coleridge Mark McGuinness reads and discusses a passage from ‘The Rime of the Ancient Mariner' by Samuel Taylor Coleridge.Poet Samuel Taylor ColeridgeReading and commentary by Mark McGuinnessFrom...

In this episode, Aydin sits down with Paul Xue, a self-described “vibe marketer” and former 3x CTO who now runs an AI-native Reddit growth agency. Paul explains why he believes any assumption you made about AI even three months ago is probably wrong today, and how that realization pushed him to pivot away from writing code as a long-term career.He walks through how his team ships production software where ~100% of the code is AI-generated, why 80% of the work now lives in planning and system design, and how new models like Claude Opus 4.5 and Gemini 3 let him literally “go for a walk” while his tools implement features. Along the way, Paul shares real numbers (two years of work vs 10–15 hours), what this means for agencies and devs, how he hires in an AI-native world, and gives a behind-the-scenes tour of the multi-agent workflows powering his Reddit content engine.Timestamps0:00 – Introduction1:01 – What a “vibe marketer” is and why Reddit is a power channel in the LLM era3:01 – From 3x CTO to Reddit-first entrepreneur: deciding coding isn't future-proof4:06 – GPT-3.5 + end of zero interest rates: when dev agency contracts fell off a cliff6:28 – Adoption curves: senior devs who still don't use AI and why personality matters7:57 – Running an AI-native shop where ~100% of production code is AI-generated9:48 – Two years vs 10–15 hours: Paul's personal 10x story on shipping an MVP12:04 – New development workflow: “plan mode” and spending 80% of time on specs18:17 – Claude Opus 4.5, Gemini 3, and “going for a walk” while AI finishes features23:30 – How $60K–$250K apps turn into weekend side projects with vibe coding tools27:12 – Hiring in the AI era: why pure “ticket-taking” devs won't survive35:12 – Inside an AI-native Reddit engine: n8n workflows, agents, Pinecone & OpenRouterTools & Technologies MentionedReddit – Primary growth and content channel; a highly trusted source for LLM training and citations.ChatGPT / GPT-3.5 – Early model that triggered Paul's realization that traditional coding careers would change.Claude 3.5 Sonnet & Claude 3.5 Opus / Opus 4.5 – Anthropic models Paul uses for long-running coding, planning, and browser automation.Gemini 3 – Google model Paul uses to quickly generate solid, familiar SaaS-style UI/UX ideas.Cursor – AI-native code editor that turns detailed “plans” into production code with one click.n8n – Automation platform that powers Paul's multi-step AI workflows for content creation and evaluation.Pinecone – Vector database storing each client's knowledge base for highly relevant Reddit responses.OpenRouter – Routing layer that lets Paul easily swap and test different language models over time.MCP (Model Context Protocol) – Framework he uses to give agents tool access (e.g., scraping Reddit, reading DBs).Notion – Fast prototyping environment to validate data models and workflows before writing custom code.Zapier – General automation glue in the earliest workflow experiments.Figma – Design tool, now increasingly AI-assisted, for UI/UX mockups.SpecCode – Tool Paul cites for vibe coding HIPAA-compliant applications.Anything – Mobile-focused “vibe coding” platform for building iOS/Android apps on your phone.Fellow – AI meeting assistant that joins meetings, produces summaries/action items, and acts as an AI chief of staff.Subscribe at⁠ thisnewway.com⁠ to get the step-by-step playbooks, tools, and workflows.

In this episode, Phillis Levin reads "An Anthology of Rain," the title poem of her newest poetry collection. She guides us through the philosophical underpinnings of her poem, how it informs the book as a whole, and how the surfaces of things can tell us so much about their substance. Phillis Levin is the author of six poetry collections, including An Anthology of Rain (https://barrowstreet.org/press/product/an-anthology-of-rain-phillis-levin/). She is also the editor of The Penguin Book of the Sonnet: 500 Years of a Classic Tradition in English (https://www.penguinrandomhouse.com/books/333350/the-penguin-book-of-the-sonnet-by-various/). Levin's honors include a Fulbright Scholar Award to Slovenia, an Ingram Merrill Grant, the Richard Hugo Prize from Poetry Northwest, and fellowships from the Guggenheim Foundation, the National Endowment for the Arts, and the Trust of Amy Lowell. To learn more about Phillis and her work, please visit her website. https://phillislevin.com Photo credit: Sigrid Estrada