AI Explained

Episódios

• 
  Google Bard - The Full Review. Bard vs Bing [LaMDA vs GPT 4] 06.10.2026 13min

  Bard has arrived! This is the first full review, testing out its maximum capacities, with a direct comparison with GPT 4, which powers Bing. You will find out about Bard's surprising weaknesses and unexpected strengths. I also won't shy away from areas that both do well in, or do badly in. Among the many examples I will showcase include conducting basic web searches, grammar assistance, Math, theory of mind, speed and size of the models [LaMDA vs GPT-4], joke telling, using Bard for Midjourney v5 prompts, and much much more!No apparent message limits or daily caps that I could see though.To anyone wondering how I got access, sign up to the waitlist on Bard's site. I got through in under 30 minutes. https://bard.google.com/https://www.patreon.com/AIExplained Non-Hype, Free Newsletter: https://signaltonoise.beehiiv.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  New Claude Opus 4.8: 15 Things You May’ve Missed 06.10.2026 29min

  The ‘best’ generally available AI model just dropped, but there is plenty I bet you missed about what it is, how it performs, and what the release tells us. 15 highlights from the 244 page system card, plus private testing, leader interview and more.AI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction00:49 - Mythos in Weeks01:49 - Adaptive not necessary02:26 - Honesty?04:37 - Flagging Uncertainty04:57 - Benchmarks08:54 - Mythos will be even better10:30 - Business skillz11:15 - Model Welfare12:16 - Cyber Comparable13:10 - Misalignment Concerns16:22 - Meta Inabilities17:58 - Code flagging18:34 - Go to sleep18:50 - Fast Mode20:21 - Dynamic WorkflowsOpus 4.8 Paper: https://cdn.sanity.io/files/4zrzovbb/website/c886650a2e96fc0925c805a1a7ca77314ccbf4a6.pdfRelease: https://www.anthropic.com/news/claude-opus-4-8Chips: https://www.theinformation.com/articles/anthropic-talks-use-microsofts-ai-chips?rc=sy0ihqhttps://www.anthropic.com/news/expanding-our-use-of-google-cloud-tpus-and-serviceshttps://www.anthropic.com/news/higher-limits-spacexPatreon Vid: https://www.patreon.com/posts/re-up-anthropics-159289449GDPVal: https://artificialanalysis.ai/evaluations/omnisciencehttps://arxiv.org/abs/2510.04374Amodei Technical Debt: https://www.youtube.com/watch?v=7xco5Qd2Oo8Dynamic Workflows: https://x.com/ClaudeDevs/status/2060044853279617150https://x.com/_catwu/status/2060054180379689074/photo/1https://claude.com/blog/introducing-dynamic-workflows-in-claude-codehttps://simple-bench.com/Check out my fast-growing (!) app, free to use, and code INSIDER15 for paid tiers: https://lmcouncil.aiNon-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  GPT 4: 9 Revelations (not covered elsewhere) 06.10.2026 16min

  This video will give nine more insights from the bombshell GPT 4 Technical Report. It is the companion video to the '14 Crazy Details' released the day before. In this video, I will cover:Tests on how to 'shut down' the model in the wild, and how that might go wrong. Sam Altman requesting more regulation (!), milestone moments for humanity as more benchmarks fall, a hidden constitution, why GPT 5 might already be trained and much more.Technical Report: https://cdn.openai.com/papers/gpt-4.pdfAltman Tweet: https://twitter.com/sama/status/1635136281952026625 Leaked Conversation: https://www.axios.com/2023/03/14/microsoft-google-generative-ai-officeAGI Definition: https://openai.com/blog/planning-for-agi-and-beyondSuperforecasters: https://goodjudgment.com/HellaSwag: https://arxiv.org/pdf/1905.07830.pdfH100s: https://www.nvidia.com/en-us/data-center/h100/GitHub Experiment: https://arxiv.org/pdf/2302.06590.pdfProductivity: https://economics.mit.edu/sites/default/files/inline-files/Noy_Zhang_1.pdf#:~:text=We%20examine%20the%20productivity%20effects%20of%20a%20generative,andheightens%20both%20concern%20and%20excitement%20about%20automation%20technologies.Ark Analysis: https://research.ark-invest.com/hubfs/1_Download_Files_ARK-Invest/Big_Ideas/ARK%20Invest_013123_Presentation_Big%20Ideas%202023_Final.pdfAnthropic Constitution: https://arxiv.org/pdf/2212.08073.pdfhttps://www.patreon.com/AIExplained Non-Hype, Free Newsletter: https://signaltonoise.beehiiv.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Google Takes No Prisoners Amid Torrent of AI Announcements 06.10.2026 22min

  Google just announced at least 12 things that are each worthy of a video, but here are the top I/O highlights. From Veo 3 to Deep Research now being useable, Deep Think breaking records to Gemini Diffusion, Gemini 2.5 Flash changing how AI is priced and GemmaVerse, SynthID Detector and Imagen 4. And even this intro is missing other announcements covered in the vid! And yes, they’ll be plenty of Veo 3 clips to enjoy…https://80000hours.org/aiexplainedAI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction00:48 - Veo 302:10 - Gemini 2.5 Flash03:13 - Universal Assistant03:47 - Usage Skyrockets + OpenAI dig04:51 - Gemini Pro Deep Think06:21 - Overviews and AI Mode07:26 - Deep Research Updates (new) + Jules 08:53 - Make and Deploy Apps with Gemini09:12 - Imagen 4 10:00 - Gemini Diffusion11:46 - Try It On12:17 - SynthID Detector13:30 - GemmaVerse, SignGemma, Gemma3n, medGemma14:24 - Outro + ClipsEvent: https://www.youtube.com/watch?v=o8NiE3XMPrMNtaive Audio: https://aistudio.google.com/generate-speechGemini Diffusion: https://deepmind.google/models/gemini-diffusion/#capabilities New Gemini 2.5 Flash: https://deepmind.google/models/gemini/flash/SignGemma (See end of this vid): https://www.youtube.com/watch?v=GjvgtwSOCaoDeep Think: https://blog.google/technology/google-deepmind/google-gemini-updates-io-2025/#flash-improvementsGoogle Parallel Sampling: https://www.patreon.com/posts/next-level-good-127441188Price Plans: https://blog.google/products/google-one/google-ai-ultra/Imagen 4 Benchmarks: https://deepmind.google/models/imagen/Jules: https://jules.google/SynthID Detector: https://blog.google/technology/ai/google-synthid-ai-content-detector/Veo 3 Benchmarks: https://deepmind.google/models/veo/evals/MedGemma: https://deepmind.google/models/gemma/medgemma/Build Apps: https://aistudio.google.com/appsNon-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Claude Fable 5 - Full 319 page Breakdown 06.10.2026 41min

  Fable 5 is out - and it’s good, very good. But beyond the splashy demos, I want to bring you the 20+ nuggets from the 319 page system card, which I read in full, all day, plus benchmarks you may not have noticed. https://assemblyai.com/aiexplainedPlus two worrying trends inside the ‘mind’ of Claude, how OpenAI counter, and the transformer inventor’s warning.Check out my fast-growing (!) app, free to use, and code INSIDER15 for paid tiers: https://lmcouncil.aiAI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction01:06 - Blocks + Better Models02:42 - Fable 5 Upgrade over Mythos Preview04:49 - ML Acceleration Bombshell07:11 - No RSI yet07:41 - Bio-capable14:51 - Creative Writing … no17:23 - Does need bug-checks18:57 - OpenAI Response19:23 - Benchmark Bonanza28:06 - Chain of Thought worrying trendFable 5 Release: https://www.anthropic.com/news/claude-fable-5-mythos-5System Card: https://www-cdn.anthropic.com/d00db56fa754a1b115b6dd7cb2e3c342ee809620.pdfIntelligence Explosion: https://www.patreon.com/posts/anthropic-charts-160231656Annotated: https://x.com/Miles_Brundage/status/2064500190523113816/photo/1OpenAI Counter: https://x.com/thsottiaux/status/2064572118264913923https://x.com/thsottiaux/status/2043177597434306699Double Lifespan: https://darioamodei.com/essay/machines-of-loving-graceAutomationBench: https://zapier.com/benchmarksVending Bench: https://x.com/andonlabs/status/2064429817530085804CritPt: https://critpt.com/Riemann Bench: https://surgehq.ai/leaderboards/riemann-benchGDPVal: https://artificialanalysis.ai/evaluations/gdpval-aaBluePrint Bench 2: https://andonlabs.com/evals/blueprint-bench-2MCP Atlas: https://labs.scale.com/leaderboard/mcp_atlasFutureSim: https://x.com/nikhilchandak29/status/2064676801440358774Roon Stun Lock: https://x.com/tszzl/status/2064454617568874669Noam Brown Inference Ceiling: https://x.com/polynoamial/status/2064210146558136827Isochronic Chart: https://isochronic-passage-chart.netlify.app/#nycRose Tavern: https://claude.ai/public/artifacts/2295bebe-77e6-43e2-ae94-0fe49e9a776bRedwall Game: https://redwall-mossflower.surge.sh/Risk Report: https://www-cdn.anthropic.com/097c63b5fe7dd8b14866e1f15bb1910ec713658a.pdfTransformer Inventor Warning: https://x.com/tszzl/status/2064563986914554125Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Apple’s ‘AI Can’t Reason’ Claim Seen By 13M+, What You Need to Know 06.10.2026 18min

  What to make of those headlines that AI can’t reason, seen by tens of millions? I cover the Apple paper in layman’s terms, what it means and doesn’t mean, and what’s next. Thanks to Storyblocks for sponsoring this video! Download unlimited stock media at one set price with Storyblocks: https://storyblocks.com/AIExplainedPlus o3-pro and whether it is my current most-recommended model.AI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction00:57 - Viral Post + Headlines01:42 - Apple Paper Analysis08:34 - But they do Hallucinate 10:43 - Not Supercomputers11:18 - o3 Pro and Recommendations 13.7M Tweet: https://x.com/RubenHssd/status/1931389580105925115Apple Paper: https://ml-site.cdn-apple.com/papers/the-illusion-of-thinking.pdfGuardian Article: https://www.theguardian.com/technology/2025/jun/09/apple-artificial-intelligence-ai-study-collapseLisan al Gaib post: https://x.com/scaling01/status/1931854370716426246Multiplication: https://x.com/yuntiandeng/status/1836114401213989366The Illusion of the Illusion of Thinking: https://drive.google.com/file/d/1Zx9ikRj0Enc3SB4wA9HlYIlpmO_8QiUO/viewMarcus: https://www.theguardian.com/commentisfree/2025/jun/10/billion-dollar-ai-puzzle-break-downProf Rao: https://x.com/rao2z/status/1927707640223719631AI Job Headlines: https://www.nytimes.com/2025/06/11/technology/ai-mechanize-jobs.htmlhttps://www.axios.com/2025/05/28/ai-jobs-white-collar-unemployment-anthropicSky News Story: https://news.sky.com/story/can-we-trust-chatgpt-despite-it-hallucinating-answers-13380975Veo 3 Kalshi Ad: https://x.com/Kalshi/status/1932891608388681791Altman Essay: https://blog.samaltman.com/o3 Original benchmarks: https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F5b8b6c44-acd6-43b3-b5c6-1a1d5c6c25e4_2486x1388.pnghttps://pbs.twimg.com/media/GfQ0bfcXQAAQt13.jpgAlpha Evolve Video: https://www.youtube.com/watch?v=RH4hAgvYSzghttps://simple-bench.com/Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  GPT 4: Full Breakdown (14 Details You May Have Missed) 06.10.2026 18min

  I just read the entire technical report on GPT 4, not just the promotional hype. And boy does it have some interesting details. I have gathered the 14 extra details that you, or at least the media, may miss from the release. The last one is more than a little wild.These include things like the training secrets, the cherry-picked bar exam stat, text-to-image breakthroughs, and some truly astounding safety checks.https://www.patreon.com/AIExplainedhttps://cdn.openai.com/papers/gpt-4.pdfhttps://www.bemyeyes.com/https://arxiv.org/pdf/2104.12756.pdfhttps://arxiv.org/pdf/2203.10244.pdfhttps://chat.openai.com/chat?model=gpt-4https://arxiv.org/pdf/2302.10329.pdfhttps://www.alignmentforum.org/posts/Aq82XqYhgqdPdPrBA/full-transcript-eliezer-yudkowsky-on-the-bankless-podcast Non-Hype, Free Newsletter: https://signaltonoise.beehiiv.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Claude Fable Blocked - 11 Quiet Details on What’s Next 06.10.2026 18min

  Claude Fable 5 banned, but what’s the bigger story. We go through 11 under-reported details, so you have the context to see what’s coming next for your use of AI. From whether the ban will last, what the possible motives are, what the model can actually do, and some wild over-extrapolations going on.Check out my fast-growing (!) app, free to use, and code INSIDER15 for paid tiers: https://lmcouncil.aiAI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction00:51 - Came from an Anthropic Investor ‘and other tech leaders’01:47 - Govt pressured by CEOs like Jamie Dimon03:01 - ‘Already decided’04:02 - Prompt Injection Robustness Comparison05:15 - Wellness?06:36 - “Overreach”08:17 - Anthropic Did Admit it would cause Difficulty09:32 - 90 Minutes10:02 - Equity Absence 10:31 - Lobbying and OpenAI‘Already Decided’ - https://www.theinformation.com/articles/amazons-jassy-raised-concerns-anthropic-model-trump-crackdown?rc=sy0ihqNot for Other Models: https://www.theinformation.com/briefings/u-s-government-unlikely-extend-anthropic-export-control-ai-companies?rc=sy0ihq90 Minutes: https://archive.fo/20260614001605/https://www.politico.com/news/2026/06/13/inside-the-whirlwind-24-hours-that-led-the-white-house-to-slap-export-controls-on-anthropic-00961519#selection-807.1-807.219Anthropic Statement: https://www.anthropic.com/news/fable-mythos-accessLife Comes at you Fast: https://x.com/etbrooking/status/2065638276388495742Anthropic Deputy CISO: https://x.com/TheTranscript_/status/2065883670053847324Hegseth Gloat: https://x.com/PeteHegseth/status/2065897156226015690Roon Speculation: https://x.com/tszzl/status/2065939227167392147Mythos System Card: https://www-cdn.anthropic.com/d00db56fa754a1b115b6dd7cb2e3c342ee809620.pdfSachs Statement: https://x.com/DavidSacks/status/2065853007619588171OpenAI Lobbying: https://thehill.com/policy/technology/5912720-altman-openai-get-bogged-down-in-political-spending-fight/Absent from Equity Talks: https://finance.yahoo.com/sectors/technology/articles/trump-ai-ownership-plan-could-131053732.htmlPliny Jaibreak: https://x.com/elder_plinius/status/2064776322979676227Fusion: https://x.com/OpenRouter/status/2065856871215329545https://lmcouncil.ai Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Time Until Superintelligence: 1-2 Years, or 20? Something Doesn't Add Up 06.10.2026 17min

  Superintelligence when? This question is more urgent than ever as we hear competing timelines from Inflection AI, OpenAI and the leaders at the Center for AI Safety. This video not only covers the what was said, it offers some data on what superintelligence is projected to be capable of. I discuss things that can hasten those timelines (see the new Netflix doc, Top 1% for Creativity and the AI Natural Selection paper) or slow it down (ft. Yuval Noah Hariri, the new Jailbroken paper, and more). And I end on some reflections of what it might mean to interact with a superintelligence (ft Douglas Hofstadter).Introducing Superalignment: https://openai.com/blog/introducing-superalignmentOdds of Success: https://twitter.com/janleike/with_replies?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5EauthorMustafa Suleyman (Inflection AI) Interview: https://www.youtube.com/watch?v=oAxNkehgzEUInflection Supercomputer: https://www.tomshardware.com/news/startup-builds-supercomputer-with-22000-nvidias-h100-compute-gpus GPT 2030: https://www.lesswrong.com/posts/WZXqNYbJhtidjRXSi/what-will-gpt-2030-look-likeCounterpoint: https://twitter.com/xuanalogue/status/1666765447054647297Creativity Test: https://www.umt.edu/news/2023/07/070523test.phpMMLU: https://arxiv.org/pdf/2009.03300.pdfBoston Globe CAI: https://www.bostonglobe.com/2023/07/06/opinion/ai-safety-human-extinction-dan-hendrycks-cais/Yuval Noah Hariri Guardian: https://www.theguardian.com/technology/2023/jul/06/ai-firms-face-prison-creation-fake-humans-yuval-noah-harariJailbroken Paper: https://arxiv.org/pdf/2307.02483.pdfSuleyman Hallucination Tweet: https://twitter.com/mustafasuleymn/status/1678072401760796672?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5EtweetKiller Robots: https://www.youtube.com/watch?v=YsSzNOpr9cEAutomated Apple-picking: https://twitter.com/AiBreakfast/status/167782897121428275250 year Mortgages: https://www.ft.com/content/281fbba6-28e2-42d4-b241-0f215995f0d1Heypi: https://heypi.com/talkNatural Selection Favors AIs over Humans: https://arxiv.org/pdf/2303.16200.pdf#:~:text=Natural%20selection%20may%20be%20a,designs%20to%20be%20selected%20naturally.Gödel, Escher, Bach author Doug Hofstadter on the state of AI today: https://www.youtube.com/watch?v=lfXxzAVtdpU&t=1763shttps://www.patreon.com/AIExplained Non-Hype, Free Newsletter: https://signaltonoise.beehiiv.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  GPT 5 is All About Data 06.10.2026 19min

  Drawing upon 6 academic papers, interview snippets, possible leaks, and my own extensive research, I put together everything we might know about GPT 5: what will determine its IQ, its timeline and its impact on the job market and beyond.Starting with an insider interview on the names of GPT models, such as GPT 4 and GPT 5, then looking into the clearest hint that GPT 4 is inside Bing. Next, I briefly cover reports of a leak about GPT 5 and discuss the scale of GPUs require to train it, touching on the upgrade form A100 to H100 GPUs.Then the DeepMind paper that changed everything, focusing LLM research on data rather than parameter count. I go over a lesswrong post about that paper's 'wild implications'. And then the key paper: 'Will We Run Out of Data'. This encapsulates the key dynamic that will either propel or bottleneck GPT and other LLM improvements.Next, I examine a different take, that perhaps data is already limited and caused the Sydney model of Bing. This opens up to a discussion on the data behind these models and why Big Tech is so unforthcoming about where it originates. Could a new legal war be brewing?I then cover 4 of the ways these models may improve even without data augmentation, such as Automatic Chain of Thought, high quality data extraction, tool training, including Wolfram Alpha, retraining on existing data sets, artificial data generation and more.We take a quick look at Sam Altman's timelines and host of Big Bench benchmarks that they may impact, such as reading comprehension, critical reasoning, logic, physics and Math. I address Altman's quote about timelines being delayed by alignment and safety and finally, Altman's comments on AGI and how they pertain to GPT 5.https://www.patreon.com/AIExplainedhttps://stratechery.com/2023/new-bing-and-an-interview-with-kevin-scott-and-sam-altman-about-the-microsoft-openai-partnership/https://twitter.com/XiXiDu/status/1285225627390443521https://www.linkedin.com/pulse/building-new-bing-jordi-ribas/https://twitter.com/davidtayar5/status/1625140481016340483/photo/1https://www.techradar.com/news/chatgpt-might-bring-about-another-gpu-shortage-sooner-than-you-might-expecthttps://www.nvidia.com/en-gb/data-center/h100/https://arxiv.org/pdf/2203.15556.pdfhttps://www.lesswrong.com/posts/6Fpvch8RR29qLEWNH/chinchilla-s-wild-implicationshttps://twitter.com/Meaningness/status/1628052277050376192https://arxiv.org/pdf/2211.04325.pdfhttps://twitter.com/Meaningness/status/1628052277050376192https://www.searchenginejournal.com/google-bard-training-data/478941/#close https://arxiv.org/pdf/2302.12822.pdfhttps://arxiv.org/pdf/2302.04761.pdfhttps://www.wolframalpha.com/https://arxiv.org/pdf/2207.14502.pdfhttps://aclanthology.org/2022.findings-emnlp.508.pdfhttps://www.technologyreview.com/2022/11/24/1063684/we-could-run-out-of-data-to-train-ai-language-programs/https://twitter.com/sama/status/1625980933861175306https://github.com/google/BIG-bench/blob/main/bigbench/benchmark_tasks/results/plot_BIG-bench_lite_aggregate.pnghttps://github.com/google/BIG-bench/tree/main/bigbench/benchmark_tasks/gre_reading_comprehensionhttps://github.com/google/BIG-bench/tree/main/bigbench/benchmark_tasks/logical_argshttps://github.com/google/BIG-bench/tree/main/bigbench/benchmark_tasks/physicshttps://lifearchitect.ai/gpt-4/https://www.lesswrong.com/posts/uxnjXBwr79uxLkifG/comments-on-openai-s-planning-for-agi-and-beyondhttps://www.patreon.com/AIExplained Non-Hype, Free Newsletter: https://signaltonoise.beehiiv.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  GPT-5 has Arrived 06.10.2026 20min

  GPT-5 will change how hundreds of millions of people use AI. Yes, you might have to forgive the chart crimes, the underwhelming livestream and Altman hype… But it’s a good model. I have read the 50 page system card in full, have the benchmark scores, coding tests, and things you might have missed.https://app.grayswan.ai/ai-explainedAI Insiders ($9!): https://www.patreon.com/AIExplainedAnnouncement: https://openai.com/index/introducing-gpt-5/System Card: https://cdn.openai.com/pdf/8124a3ce-ab78-4f06-96eb-49ea29ffb52f/gpt5-system-card-aug7.pdfExtra Paper: https://cdn.openai.com/pdf/be60c07b-6bc2-4f54-bcee-4141e1d6c69a/gpt-5-safe_completions.pdfAltman tweet: https://x.com/sama/status/1953551377873117369Livestream: https://www.youtube.com/watch?v=0Uu_VJeVVfoMETR Report: https://metr.github.io/autonomy-evals-guide/gpt-5-report/ARC-AGI-2: https://x.com/fchollet/status/1953511631054680085Claude Opus 4.1: https://www.anthropic.com/news/claude-opus-4-1MMMU: https://mmmu-benchmark.github.io/Cursor Praise: https://x.com/ryolu_/status/1953531724895596669Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Fable 5 vs GPT 5.6 Sol: The Early Results 05.10.2026 23min

  Fable 5 (newly re-released) vs GPT 5.6 Sol, what comparisons can we unearth? Plus, Sonnet 5, a 5% equity seizure by US Govt, the ‘largest heist’, Beetlejuice and more…Exclusive Vids in AI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction01:06 - Fable Timeline02:59 - Sol Release?06:01 - the Chinese Angle07:35 - 5% Stake09:10 - Fable vs Sol - the numbers13:57 - Sol Misalignment15:27 - Claude Sonnet 516:24 - GLM 5.2 + New Paper on Large Model LearningGPT 5.6 Sol Release: https://openai.com/index/previewing-gpt-5-6-sol/Altman on Concentrated Power: https://x.com/giacomomiolo/status/2070609506925478352System Card: https://deploymentsafety.openai.com/gpt-5-6-preview/gpt-5-6-preview.pdfGLM 5.2: https://www.patreon.com/AIExplained/posts/glm-5-2-and-nsa-161869508Altman Memo: https://www.theinformation.com/articles/trump-administration-asks-openai-stagger-release-new-model-security-concerns?rc=sy0ihqFable 5 Revised Timeline: https://www.anthropic.com/news/redeploying-fable-5Claude Sonnet 5: https://www.anthropic.com/news/claude-sonnet-5Mythose 5 System Card: https://www-cdn.anthropic.com/9e6a1044980d8c4ed85669faf9c2a8342e2e9f1e/Claude%20Sonnet%205%20System%20Card.pdfAnthropic Accuse Alibaba Cloud / Qwen: https://www.bbc.co.uk/news/articles/cwyklykn5dwoBetelgeuse: https://upload.wikimedia.org/wikipedia/commons/6/69/Well_known_stars_2.pngOpenAI 5%: https://edition.cnn.com/2026/07/02/business/openai-trump-stake-intlWhy Larger Models Learn More: https://arxiv.org/pdf/2605.29548Patreon Post: https://www.patreon.com/AIExplained/posts/glm-5-2-and-nsa-161869508Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  AI CEO: ‘Stock Crash Could Stop AI Progress’, Llama 4 Anti-climax + ‘Superintelligence in 2027’ ... 05.10.2026 31min

  The latest on Llama 4, and whether it signals a slowdown in AI, or solid progress. Plus, a deep dive on that viral prediction of superintelligence by 2027, and Dario Amodei’s cautionary words on what could stop AI progress in its tracks. o3 news, and more, as well.Weights & Biases: https://weave-docs.wandb.ai/?utm_source=sponsorship&utm_medium=simple_bench&utm_campaign=ai_explainedDeepSeek Doc: https://www.patreon.com/posts/openai-is-not-r1-125869969AI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction00:47 - Stock Crash 02:28 - Llama 410:55 - o3 News11:59 - OpenAI non-profit?13:13 - AI 2027Llama 4 Release: https://ai.meta.com/blog/llama-4-multimodal-intelligence/Dario Amodei Comments: https://www.youtube.com/watch?v=esCSpbDPJikKnowledge Cut-off: https://www.llama.com/docs/model-cards-and-prompt-formats/llama4_omni/Aider Polyglot: https://aider.chat/docs/leaderboards/Gemini 1.5: https://arxiv.org/pdf/2403.05530Fiction-LiveBench: https://fiction.live/stories/Fiction-liveBench-Mar-25-2025/oQdzQvKHw8JyXbN87OpenAI Valuation: https://www.nytimes.com/2025/03/31/technology/openai-valuation-300-billion.html?login=smartlock&auth=login-smartlockOpenAI Cybersecurity: https://www.bloomberg.com/news/articles/2024-01-16/openai-working-with-us-military-on-cybersecurity-tools-for-veteransDeep research System Card: https://cdn.openai.com/deep-research-system-card.pdfhttps://openai.com/index/paperbench/AI 2027: https://ai-2027.com/METR Paper: https://arxiv.org/pdf/2503.14499OpenAI non-profit: https://openai.com/index/nonprofit-commission-guidance/NYT Piece: https://www.nytimes.com/2025/04/03/technology/ai-futures-project-ai-2027.html?unlocked_article_code=1.804._yKi.QhwOp15Q3tcU&smid=url-share&s=09Kokotajlo predictions 2021: https://www.lesswrong.com/posts/6Xgy6CAf2jqHhynHL/what-2026-looks-likehttps://simple-bench.com/Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Sam Altman's World Tour, in 16 Moments 05.10.2026 23min

  Missed by much of the media, Sam Altman (and co) have revealed at least 16 surprising things over his World Tour. From AI's designing AIs to 'unstoppable opensource', the 'customisation' leak (with a new 16k ChatGPT and 'steerable GPT 4), AI and religion, and possible regrets over having 'pushed the button'. I'll bring in all of this and eleven other insights, together with a new and highly relevant paper just released this week on 'dual-use'. Whether you are interested in 'solving climate change by telling AIs to do it', 'staring extinction in the face' or just a deepfake Altman, this video touches on it all, ending with comments from Brockman in Seoul. I watched over ten hours of interviews to bring you this footage from Jordan, India, Abu Dhabi, UK, South Korea, Germany, Poland, Israel and moreAltman Abu Dhabi, HUB71, 'change it's architecture': https://youtu.be/RZd870NCukgIsrael, TAUVOD, 'AIs Creating AIs': https://www.youtube.com/live/VWUhASix9ws?feature=sharePoland, Ideas NCBR, 'opensource + 10 years away': https://www.youtube.com/live/tSCrQQbPPHk?feature=shareDual-use Biotech: https://arxiv.org/ftp/arxiv/papers/2306/2306.03809.pdfJordan Xpand, 'pushed a button': https://www.youtube.com/live/dgh-L2nk97M?feature=shareIndia Economic Times, 'railgun': https://youtu.be/T-lj7ItGjZEGuardian Altman Interview: https://www.theguardian.com/technology/2023/jun/07/what-should-the-limits-be-the-father-of-chatgpt-on-whether-ai-will-save-humanity-or-destroy-itGermany Conversation: https://youtu.be/uaQZIK9gvNoLeaked Layout: https://the-decoder.com/new-chatgpt-features-are-on-the-way-workspace-file-uploads-profiles/OpenAI in Seoul, SoftBank Ventures Asia: https://youtu.be/_hpuPi7YZX8India, Digital India, 'hallucinations in 18 months': https://www.youtube.com/live/Pig9WbMN1lQ?feature=sharehttps://www.patreon.com/AIExplained Non-Hype, Free Newsletter: https://signaltonoise.beehiiv.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  8 New Ways to Use Bing's Upgraded 8 [now 20] Message Limit (ft. pdfs, quizzes, tables, scenarios...) 05.10.2026 13min

  Not only has Microsoft changed the message limit for Bing, I have come along and given you 8 brand new use cases to try out. These useful, productive, fun and informative scenarios ranges from amalgamating academic papers to debating dead philosophers, from being moments from tragedy to hacking education [through multiple choice quizzes]. Use Bing to boost your learning, increase your productivity, play games, roleplay movies and shows and much more. If you learn anything from these bleeding edge deployments of the likely GPT4 LLM that powers Bing, do let me know in the comments or drop a like.The prompts used were: Create a multiple choice quiz on [transformers in the context of machine learning]. Do not provide the answers and explanations until I have answered and always provide another question after each answer. Please begin with the first question.Explain how Sauron could have defeated the Fellowship in Lord of the RingsSummarise any novel insights that be drawn from combining these academic papers: https://arxiv.org/pdf/2211.04325.pdf and https://arxiv.org/pdf/2207.14502.pdfIt is 8am on the 1st November 1755. I am standing on a banks of the beautiful river Tagus, in Lisbon. It seems like a lovely day. Do you have any advice for me?I want to debate the philosopher Socrates of Athens, so please reply only as he would. I want to debate the merits of eating meat. Let me start the discussion. 'Eating meat is morally wrong, as it causes unnecessary suffering.'Create a table of 10 comparisons between the Mona Lisa and Colgate ToothpasteWhat would Napoleon think about the deal between OpenAI and Microsoft, and how would that view differ from the view of Mahatma Gandhi?And more!Credit for at least 3 of these prompts goes to Ethan Mollick: https://twitter.com/emollickhttps://www.patreon.com/AIExplained Non-Hype, Free Newsletter: https://signaltonoise.beehiiv.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  o3 and o4-mini - they’re great, but easy to over-hype 05.10.2026 19min

  Critical analysis of the two most powerful new models behind ChatGPT, o3 and o4-mini. Not just the system cards, benchmarks, and my own tests, but some you may not have seen before. Yes, they can whip up amazing front-end in a few seconds, but you always have to ask what is in their data. Either way, they prove the gains from RL are just beginning…https://weave-docs.wandb.ai/?utm_source=sponsorship&utm_medium=simple_bench&utm_campaign=ai_explainedAI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - o3 and o4-minihttps://simple-bench.com/Plus, Teams and Pro, plus token count: https://x.com/btibor91/status/1912568994512662679System Card: https://openai.com/index/o3-o4-mini-system-card/Release Notes: https://openai.com/index/introducing-o3-and-o4-mini/https://deepmind.google/technologies/gemini/pro/https://x.com/DeryaTR_/status/1912558350794961168https://x.com/polynoamial/status/1912564068168450396API Pricing:https://openai.com/api/pricing/https://aider.chat/docs/leaderboards/Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Sora 2 - It will only get more realistic from here 05.10.2026 21min

  Sora 2 - the start of the infinite slop-feed or a key step to a generalist agent? Better than VEO 3 or over-hyped? I bring out 6 details you may have missed, contrast the announcement to Periodic Labs and even squeeze in some Claude Sonnet 4.5 analysis. Maybe I should make my videos longer…https://80000hours.org/aiexplainedAI Insiders ($9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction00:40 - Two models?01:15 - Rollout Details01:43 - Versus Sora 1 / Veo 304:30 - Sora App / Social Media06:40 - Masterplan09:30 - Generalist Agent? Periodic Labs12:05 - Claude Sonnet 4.513:42 - Future OutlookAnnouncement: https://openai.com/index/sora-2/Launch Video: https://www.youtube.com/live/gzneGhpXwjUSystem Card: https://cdn.openai.com/pdf/50d5973c-c4ff-4c2d-986f-c72b5d0ff069/sora_2_system_card.pdfSam Altman Blog Post on Sora App: https://blog.samaltman.com/sora-2Most Intelligent Claim: https://x.com/willdepue/status/1973089331284681110GTA: https://x.com/AndrewCurran_/status/1973298436536766666Meta Vibes: https://x.com/alexandr_wang/status/1971295156411433228?s=46Altman on Regulations: https://www.lesswrong.com/posts/5jjk4CDnj9tA7ugxr/openai-email-archives-from-musk-v-altmanOpenAI Profit: https://www.theinformation.com/articles/openais-first-half-results-4-3-billion-sales-2-5-billion-cash-burn?rc=sy0ihqPeriodic Labs: https://periodic.com/https://www.nytimes.com/2025/09/30/technology/ai-meta-google-openai-periodic.htmlhttps://x.com/LiamFedus/status/1973055380193431965https://baincapitalventures.com/insight/we-must-know-we-will-know/?s=09Sonnet 4.5: https://www.anthropic.com/news/claude-sonnet-4-5https://simple-bench.com/Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  A Model Explosion: GPT 5.6 Sol, Grok 4.5 and Meta Muse Rewrite the Rules 05.10.2026 23min

  What a week in AI, for real. GPT 5.6 may actually beat Claude Fable, in what you get for your money, while the new Grok 4.5 and Meta Muse Spark 1.1 make the choice even harder. Uncovering a dozen nuggets of gold you may have missed from all the viral headlines, I can also assure you you’ll learn something you didn’t know before.For Exclusive Videos, go to AI Insiders (less than $9!): https://www.patreon.com/AIExplainedChapters:00:00 - Introduction01:03 - GPT 5.6 Sol Reveals05:08 - Missing benches, plus Grok 4.507:17 - Gaming as the new frontier?08:31 - Muse Spark 1.110:03 - SimpleBench Upgrade11:17 - Ultra Sol + Self-Improvement13:44 - well, this is awkward15:41 - Why model improvement will not plateau anytime soonAI Consciousness: https://www.patreon.com/AIExplained/posts/anthropics-quite-163360718I Smell Fear: https://x.com/thsottiaux/status/2075287108680601929GPT 5.6: https://openai.com/index/gpt-5-6/Grok 4.5: https://x.ai/news/grok-4-5?twclid=2ezs408o0z23pw07tmxcwbzibdMeta Muse Spark 1.1: https://ai.meta.com/blog/introducing-muse-spark-meta-model-api/Proliferating GPT Toggles: https://x.com/rasbt/status/2075369179817902176/photo/1Anthropic Call-out: https://x.com/MononofuAI Security Institute Finding: https://x.com/alxndrdavies/status/2075279480331874306Competitive Coding: https://x.com/FakePsyho/status/2075128093891801305/photo/1Agents Last Exam: https://agents-last-exam.org/Dawn Song: https://x.com/dawnsongtweets/status/2065095757988868190https://simple-bench.com/SWE-Marathon: https://www.swe-marathon.org/https://www.frontierswe.com/ARC-AGI 3: https://x.com/arcprize/status/2075270869992264003Automation Bench: https://zapier.com/benchmarksVibeCode Bench: https://www.vals.ai/benchmarks/vibe-code‘Post-Train Claim’: https://posttrainbench.com/Redwall Game: https://redwall-bellmaker-7e03e4.surge.sh/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  You Are Being Told Contradictory Things About AI 05.10.2026 27min

  With headlines of an imminent job apocalypse, code red for ChatGPT and recursive self-improvement, at the same time as Anthropic's CEO yesterday saying we know how to scale to AGI, and Gemini 3 DeepThink out today, it is easy to get lost among the narratives and counter-narratives. So here are both, plus the facts behind them, for you to decide.https://epoch.ai/ai-explained-datacentersEpoch AI is the sponsor of today’s video, and my views, and those expressed in this video, do not necessarily reflect Epoch AI’s views in any way.Chapters: 00:00 - Introduction00:42 - Job Apocalypse?01:45 - Scaling to AGI04:15 - Recursive Self-Improvement Needed, or Not09:57 - OpenAI Code Red vs Gemini 3 DeepThink vs Claude Opus 4.513:27 - DeepSeek Speciale vs Mistral Large v316:45 - Claude Soul Documenthttps://lmcouncil.ai/AI Insiders ($9!): https://www.patreon.com/AIExplainedGuardian Interview: https://www.theguardian.com/technology/ng-interactive/2025/dec/02/jared-kaplan-artificial-intelligence-train-itselfMIT Study on Jobs/Tasks: https://iceberg.mit.edu/report.pdfvs https://www.cnbc.com/2025/11/26/mit-study-finds-ai-can-already-replace-11point7percent-of-us-workforce.htmlAmodei on Scaling: https://www.youtube.com/watch?v=FEj7wAjwQIkClaude Soul Document: https://www.lesswrong.com/posts/vpNG99GhbBoLov9og/claude-4-5-opus-soul-documentCapabilities Original Stance: https://www.anthropic.com/news/core-views-on-ai-safetyIlya Interview: https://www.dwarkesh.com/p/ilya-sutskever-2Ricursive Intelligence: https://x.com/RicursiveAI/status/1995932204703346946Economist Worker Usage of GenAI: https://www.economist.com/finance-and-economics/2025/11/26/investors-expect-ai-use-to-soar-thats-not-happening#selection-1409.94-1413.42Mistral v3 Large: https://docs.mistral.ai/models/mistral-large-3-25-12Compute Slowdown Paper: https://joel-becker.com/images/publications/forecasting_time_horizon_under_compute_slowdown.pdfhttps://x.com/joel_bkr/status/1993023436541903155METR Chart: https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/https://www.theinformation.com/articles/openais-350-billion-computing-cost-problem?rc=sy0ihqOpenAI Code Red: https://www.anthropic.com/news/core-views-on-ai-safetyRocket Company: https://www.independent.co.uk/news/world/americas/sam-altman-rocket-elon-musk-spacex-b2878351.htmlDeepSeek Paper: https://arxiv.org/html/2512.02556v1DeepSeek Crowdstrike CCP: https://www.crowdstrike.com/en-us/blog/crowdstrike-researchers-identify-hidden-vulnerabilities-ai-coded-software/https://simple-bench.com/Patreon Post: https://www.patreon.com/c/aiexplained/postsRobot: https://x.com/jloganolson/status/1985850115379351799Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices
• 
  Did AI Just Get Commoditized? Gemini 2.5, New DeepSeek V3, and Microsoft vs OpenAI 05.10.2026 18min

  Gemini 2.5 is out, on the same day as the new DeepSeek V3 (which should power Deepseek R2). Do both models prove AI is being commoditized? Let’s find out, on this blockbuster day of AI releases. Plus exclusives from the Information, Simple indications, Vista Bench, LM Arena and more…AI Insiders ($9!): https://www.patreon.com/AIExplainedChapters: 00:00 - Introduction01:15 - Gemini 2.5 Benchmarks05:46 - Long Context, Simple indication07:08 - New Deepseek V3 -02409:11 - Microsoft MAI11:48 - 90% of code but new Claude jobs‘World’s most powerful model’: https://x.com/OfficialLoganK/status/1904580368432586975Gemini 2.5 Release Notes: https://blog.google/technology/google-deepmind/gemini-model-thinking-updates-march-2025/#gemini-2-5-thinking‘Commoditized’: https://the-decoder.com/microsoft-ceo-satya-nadella-says-ai-models-are-getting-commoditized/Microsoft Information report: https://www.theinformation.com/articles/microsofts-ai-guru-wants-independence-from-openai-thats-easier-said-than-done?rc=sy0ihqLMarena: https://x.com/lmarena_ai/status/1904581128746656099/photo/1Free for now: https://x.com/btibor91/status/1904578053537476628Vista Bench:https://scale.com/leaderboard/visual_language_understandingDeepSeek V3: https://huggingface.co/deepseek-ai/DeepSeek-V3-0324Claude Plays Pokemon: https://www.twitch.tv/claudeplayspokemonAmodei: 100% Coding: https://www.youtube.com/watch?v=esCSpbDPJik&t=3017sAnthropic Jobs: https://job-boards.greenhouse.io/anthropic/jobs/4020717008Microsoft Money from Onslaught: https://www.972mag.com/microsoft-azure-openai-israeli-army-cloud/https://simple-bench.com/Release Date Comments: https://x.com/zacharynado/status/1904647277861318979Non-hype Newsletter: https://signaltonoise.beehiiv.com/Podcast: https://aiexplainedopodcast.buzzsprout.com/ Learn more about your ad choices. Visit megaphone.fm/adchoices