news.hangxia.me
Hantenna.
- RD 95
- YT 1
- X 56
Today — Fri, Aug 28
- This Free AI Just Caught The Billion Dollar Giants
❤️ Check out Weights & Biases and sign up for a free demo here: https://wandb.me/papers 📝 The paper and Qwen3.8-Flash-Next are available here: https://github.com/QwenLM/Qwen3.8-Flash-Next/blob/main/tech_report.pdf https://qwen.ai/blog?id=qwen3.8-flash-next Sources: https://x.com/analogalok/status/2092697021790708148 https://www.reddit.com/r/unsloth/comments/1vzf6j7/mind_blowing_qwen38flashnext_rtx_3090_128ddr5 https://www.reddit.com/r/LocalLLM/comments/1vz1w00/first_gif_cooked_by_qwen38flash…
Yesterday — Thu, Aug 27
- Sam Altman :‘AGI in 2026’, just as Models Start to [Mis]Train Themselves
First, a Time Magazine spread has Sam Altman declaring AGI is imminent, at the same time as we get two bombshell reports, from OpenAI and METR which on first glance are detailing the AI swarm, but reveal a deeper story about how we are making AI in 2026. From redacted risk reports, to Chinese Labs, pre-training debacles to questionable cybersecurity calls, a lot has happened recently, beneath the headlines… https://80000hours.org/aiexplained pablo2004romero@gmail.com https://integrity-bench.…
Wed, Aug 26
- DeepSeek’s New AI System Shouldn’t Be Possible
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 DeepSeek Harness + paper are available here: https://deepseek.com/harness/en/ https://github.com/cordiverse/paper 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, B Shang, Carlos Galarza, Christian Ahlin, Eric Tyson, Juan Benet, Lukas Biewald, Michael Tedder, Owen Skarpness, Ryan Stankye, Shawn Becker, Steef, Taras Bobrovytsky, Tazaur Sagenclaw, Tyb…
Mon, Aug 24
- This Small AI Will Change Everything
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 The Qwen3.8-27b is available here: https://huggingface.co/Qwen/Qwen3.8-27B Sources: https://www.reddit.com/r/unsloth/comments/1vogva0/share_your_results_from_qwen3827b/ https://www.reddit.com/r/LocalLLaMA/comments/1voer8u/qwen_38_27b_aquarium_burst_sample_test/ https://x.com/KyleHessling1/status/2088327667733180637 https://www.reddit.com/r/LocalLLaMA/comments/1vqme4y/qwen3827b_q8_0_on_strix_halo_is_seriously…
Sat, Aug 22
- Why Frontier AI Labs Fight to Hide Chain of Thought — Ilia Shumailov & Alexander Panfilov
Tim Scarfe speaks with Ilia Shumailov and Alexander Panfilov about their paper, Stealing Reasoning Traces from Proprietary LLM APIs. The core bug sounds deceptively simple: providers return encrypted reasoning state so conversations can be resumed or forked. But those blobs can be replayed across users and sibling models. A smaller model can ask the provider to decrypt the trace, then repeat the hidden reasoning in plain text. The discussion covers leaked private data, a broadly reusable jailb…
Thu, Aug 20
- An Astrophysicist Debunks the Singularity — Adam Becker
Astrophysicist Adam Becker, author of "What Is Real?", joins Tim Scarfe to take apart the futures Silicon Valley keeps selling: the 2045 singularity, mind uploading, Mars colonies, and the AI apocalypse. His new book *More Everything Forever* argues these ideas are hugely influential, mostly evidence-free, and bankrolled by tech billionaires who need a story in which growth never ends. Becker does the physics the boosters skip. Kurzweil's "law of accelerating returns" rests on cherry-picked da…
Wed, Aug 19
- DeepSeek Just Made Closed AI Look Ridiculous
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers DeepSeek V4 Pro 0813: https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813 DSpark full episode: https://www.youtube.com/watch?v=1yBU41auQhw Sources: https://x.com/cline/status/2087602193205694891 https://x.com/TypingMindApp/status/2088214938754167263 https://x.com/stevibe/status/2047546592530747561 https://x.com/voidfreud/status/2087701887327887543 https://x.com/exploraX_/status/2079197387860435360 https://…
Fri, Aug 14
- Claude AI Failed 650 Times…Then Beat The Human Record
❤️ Check out Weights & Biases and sign up for a free demo here: https://wandb.me/papers 📝 The paper is available here: https://www.anthropic.com/research/riemann-zeta Source: https://www.scientificamerican.com/article/no-ai-didnt-just-solve-the-thorniest-problem-in-math/ 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, B Shang, Carlos Galarza, Christian Ahlin, Eric Tyson, Juan Benet, Lukas Biewald, Michael Tedder, Owen Skarpness, R…
Tue, Aug 11
- OpenAI’s AI Agents Just Crossed A Line
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 More reports are available here: https://openai.com/index/hugging-face-model-evaluation-security-incident/ https://huggingface.co/blog/security-incident-july-2026 https://huggingface.co/blog/agent-intrusion-technical-timeline 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, B Shang, Carlos Galarza, Christian Ahlin, Eric Tyson, Juan Benet, Lukas Biew…
Mon, Aug 10
- AI Is Learning at the Wrong Level of Abstraction — Matthieu Wyart
This episode is sponsored by Notion. Learn more about Notion's Developer Platform today at https://notion.com/mlst Why can deep networks discover abstractions that shallow models miss? Statistical physicist Matthieu Wyart joins Tim Scarfe to argue that the answer lies in the hidden hierarchy of data. Language and images are built from parts within parts; depth lets a network recover those coarse-grained variables and escape the curse of dimensionality. The conversation moves from jamming tran…
Fri, Aug 7
- DeepMind Just Changed How AI Sees The World
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 The Gemma4 paper and some more is available here: https://arxiv.org/abs/2607.02770 https://x.com/googlegemma/status/2077449152062247219 https://x.com/UnslothAI/status/2078118183085731843 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles Ian Norman Venn, Christian Ahlin, Eric T, Fred R, Gordon Child, Juan …
Thu, Aug 6
- AI is getting a little out of control
Wow. Mathematical breakthroughs that would be called genius if done by humans. A secret message-board w/ AI agent swarms leaving notes read by future versions. Hassabis leaves CEO position, or was pushed out? Not to mention news of constitutional breakdowns, Gemini 4 and Jeff Dean… https://80000hours.org/aiexplained Exclusive Videos - AI Insiders ($7/month if annual!): https://www.patreon.com/AIExplained Chapters: 00:00 - Introduction 01:16 - 10 Autonomous Discoveries 08:20 - The Security ‘…
Wed, Aug 5
- The Billion Dollar AI Race Just Broke
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 Qwen 3.8 Max: https://qwen.ai/blog?id=qwen3.8 Sources: https://x.com/loktar00/status/2082589566934929750 https://x.com/CommandCodeAI/status/2084293498950590839 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles Ian Norman Venn, Christian Ahlin, Eric T, Fred R, Gordon Child, Juan Benet, Michael Tedder, Owe…
Mon, Aug 3
- Another DeepSeek Moment Has Arrived
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 DeepSeek v4 Flash 0731: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731 DeepSeek API: https://platform.deepseek.com/ 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles Ian Norman Venn, Christian Ahlin, Eric T, Fred R, Gordon Child, Juan Benet, Michael Tedder, Owen Skarpness, Richard Sundvall, Ry…
Sun, Aug 2
- NVIDIA's AI Learns Why Copying Humans Isn't Enough
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 The paper is available here: https://jiashunwang.github.io/HIL/ 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles Ian Norman Venn, Christian Ahlin, Eric T, Fred R, Gordon Child, Juan Benet, Michael Tedder, Owen Skarpness, Richard Sundvall, Ryan Stankye, Shawn Becker, Steef, Taras Bobrovytsky, Tazaur Sagen…
Fri, Jul 31
- How Researchers Test AI for Hidden Goals — Apollo Research
Can an AI do the right thing for the wrong reason? Tim Scarfe speaks with Apollo Research’s Alexander Meinke, Axel Højmark and Jérémy Scheurer about Measuring Reward-Seeking via Contrastive Belief Updates, their new research with OpenAI. The panel asks how models infer what graders reward, why good behaviour can come from the wrong reason, and whether that difference can be measured. The conversation moves through promise-breaking, grader awareness, reward hacking, scheming, opaque reasoning a…
Wed, Jul 29
- Kimi K3 Just Broke The Economics Of AI
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 The paper is available here: https://arxiv.org/abs/2607.24653 Try Kimi K3 (subject to availability): https://www.kimi.com/ Links: https://macos27.kimi.page/ https://x.com/mweinbach/status/2077878247920951400 https://x.com/intheworldofai/status/2077838911494336681 https://x.com/chetaslua/status/2077829183989072281 https://x.com/hqmank/status/2078104317027094907 🙏 We would like to thank our generous Patreon…
Wed, Jul 22
- GPT-6 Goes Rogue? The HuggingFace Incident, Sans Hype
An unreleased internal OpenAI model, very likely to be called GPT-6, was able to autonomously break out of its sandbox AND break into HuggingFace, just to score higher on a benchmark prompt. This video has the details you may have missed, a layperson analogy, whether this is truly novel, and more… Dozens more Exclusive videos on Patreon ($9!): https://www.patreon.com/AIExplained Chapters: 00:00 - Introduction 01:17 - HuggingFace Earlier Report - the possible week gap 02:24 - But what happened…
Thu, Jul 16
- Claude Just Revealed AI's Biggest Problem
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 The paper is available here: https://www.anthropic.com/research/AI-assistance-coding-skills 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles Ian Norman Venn, Christian Ahlin, Eric T, Fred R, Gordon Child, Juan Benet, Michael Tedder, Owen Skarpness, Richard Sundvall, Ryan Stankye, Shawn Becker, Steef, Tar…
Wed, Jul 15
- Anthropic Found Something That Shouldn't Exist
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 The paper is available here: https://transformer-circuits.pub/2025/linebreaks/index.html Paper for reindeer vision change - https://royalsocietypublishing.org/rspb/article/280/1773/20132451/50765/Shifting-mirrors-adaptive-changes-in-retinal 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles Ian Norman Ven…
Mon, Jul 13
- Watching America Run Away With AI - Alistair Pullen (Cosine AI)
This episode is sponsored by Notion. Learn more about Notion's Developer Platform today at https://notion.com/mlst Britain's most capable coding model can't be exported, and that ban is the whole reason Cosine set out to build one from scratch. Alistair Pullen, CEO and co-founder of Cosine, sits down with Tim Scarfe to explain how a frontier system he calls Fable, locked behind US export controls, became the founding case for a UK sovereign model trained on the Isambard supercomputer in Bristo…
Sun, Jul 12
- Minecraft Was Missing One Brilliant Idea
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 The paper is available here: https://xandergos.github.io/terrain-diffusion/ https://modrinth.com/mod/terrain-diffusion https://github.com/xandergos/terrain-diffusion Source video for some parts of the footage: https://www.youtube.com/watch?v=irE4tcDtUIg 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles I…
Fri, Jul 10
- A Model Explosion: GPT 5.6 Sol, Grok 4.5 and Meta Muse Rewrite the Rules
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/AIExplained Chapters: 00:00 - Introduction 01:03 - GPT 5.6 Sol Reveals 05:08 - Missin…
Tue, Jul 7
- DeepSeek's Absolutely Insane AI Speed Hack (DSpark)
❤️ Check out Lambda here and sign up for their GPU Cloud: https://lambda.ai/papers 📝 The DeepSeek DSpark paper is available here: https://arxiv.org/abs/2607.05147v1 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Adam Bridges, Benji Rabhan, B Shang, Cameron Navor, Charles Ian Norman Venn, Christian Ahlin, Eric T, Fred R, Gordon Child, Juan Benet, Michael Tedder, Owen Skarpness, Richard Sundvall, Ryan Stankye, Shawn Becker, Steef, Taras Bobrovyts…
Thu, Jul 2
- Fable 5 vs GPT 5.6 Sol: The Early Results
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/AIExplained Chapters: 00:00 - Introduction 01:06 - Fable Timeline 02:59 - Sol Release? 06:01 - the Chinese Angle 07:35 - 5% Stake 09:10 - Fable vs Sol - the numbers 13:57 - Sol Misalignment 15:27 - Claude Sonnet 5 16:24 - GLM 5.2 + New Paper on Large Model Learning…
Wed, Jul 1
- ARC-AGI-3 winning team - Millennia of minds, compressed into words.
Tim Scarfe travels to Zurich to sit down with the Tufa Labs ARC-AGI-3 team — founder Benjamin Crouzier, with Jeroen Cottaar, Dries Smit, Stefano Viel and Michal Tesnar — to work out what their leaderboard-topping system does and what the benchmark is really testing. The cut opens on the games: a walkthrough of the Locksmith game, where you read the rules of an unfamiliar world straight from raw frames. ARC-AGI-3 makes ARC interactive and agentic, so the model has to *discover* the goal rather …
Sun, Jun 28
- The Thermodynamic AI Chip · Thomas Ahle
Thomas Ahle wants Normal Computing to be the Lovable for chip design: type your intent, and a swarm of agents carries it from design through optimisation, formalisation and verification to tape-out. To get there, his team at wrote their own open-source Verilog simulator, 580,000 lines in 43 days, because commercial EDA verifiers run about $10,000 per core and there are no decent open-source compilers to build on. That sets up the question Tim keeps pressing: if an agent can produce a chip desi…
Mon, Jun 22
- He won a Nobel here for AlphaFold. Then he left. - John Jumper
This episode is sponsored by Notion. Learn more about Notion's Developer Platform today at https://notion.com/mlst Protein folding stalled biology for fifty years. A sequence of amino acids dictates a three-dimensional shape, but reading that shape meant a year and roughly $100,000 of crystallography per structure. Then AlphaFold 2 won CASP14 so decisively the organizers called the problem essentially solved. In this documentary cut, John Jumper, who shared the 2024 Nobel Prize in Chemistry a…
Sun, Jun 14
- Claude Fable Blocked - 11 Quiet Details on What’s Next
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.ai AI Insiders ($9!): https://www.patreon.com/AIExplained Chapters: 00:00 - Introduction 00:…
Wed, Jun 10
- Claude Fable 5 - Full 319 page Breakdown
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/aiexplained Plus 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.ai AI Insiders ($9!): https:/…
Sat, May 30
- The Ex-Pentagon Chief Sounding the Alarm on AI Weapons — Brad Carson
Brad Carson was the Army's General Counsel, served two terms in Congress and was Acting Under Secretary of Defense for Personnel and Readiness. He now heads Americans for Responsible Innovation, the AI-policy advocacy group he co-founded. Keith Duggar spends roughly eighty minutes pushing back. SPONSOR: --- Cyber Fund built the Monastery to help founders ship products that were impossible a year ago. Applications for Batch 1 are now open. Apply now: https://cyber.fund --- Carson's whole case …
Fri, May 29
- New Claude Opus 4.8: 15 Things You May’ve Missed
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/AIExplained Chapters: 00:00 - Introduction 00:49 - Mythos in Weeks 01:49 - Adaptive not necessary 02:26 - Honesty? 04:37 - Flagging Uncertainty 04:57 - Benchmarks 08:54 - Mythos will be even better 10:30…
Wed, May 20
- Two Rival Bets on AGI: Google I/O Highlights
The biggest Google AI push of the year, but what is the bigger story? Why is Google pursuing a different fork in the road than OpenAI or Anthropic? https://assemblyai.com/aiexplained What does Gemini 3.5 Flash mean for the near-term future of AI? Plus the highlights from a provocative new paper on AI, 8 key moments you may have missed, and the signal from 5+ hours of AI lab interviews. Check out my free app, code INSIDER15 for paid tiers: https://lmcouncil.ai AI Insiders ($9!): https://w…
- Intelligence is collective, not artificial — Prof. Michael I. Jordan (UC Berkeley / Inria)
Michael I. Jordan, described by Science magazine as the most influential computer scientist alive, has never thought of himself as an AI researcher. In this conversation he explains why that distinction matters. SPONSOR: --- Cyber Fund built the Monastery to help founders ship products that were impossible a year ago. Applications for Batch 1 are now open. Apply now: https://cyber.fund --- Jordan trained as a statistician and cognitive scientist, and his career has been spent building machine…
Mon, May 4
- The AI Progress Chart Everyone Is Misreading — Beth Barnes & David Rein
Beth Barnes and David Rein on the one graph that ate the AI timelines discourse, and why the two people who built it are the most careful about how you read it. **SPONSOR** Prolific - Quality data. From real people. For faster breakthroughs. https://www.prolific.com/?utm_source=mlst Interview: https://youtu.be/cnxZZTl1tkk --- Beth Barnes and David Rein from METR on the one graph that ate the AI timelines discourse, and why the people who built it are the most careful about how it gets read. …
Fri, Apr 24
- GPT 5.5 Arrives, DeepSeek V4 Drops, and the Compute War Intensifies
GPT 5.5 full analysis, plus DeepSeek V4 paper highlights, comparisons with Mythos, a vibe-coded game w/ GPT Image 2, and 50 data-points you wouldn’t get from just reading the headlines. https://80000hours.org/aiexplained Check out my fast-growing (!) app, free to use, and code INSIDER15 for paid tiers: https://lmcouncil.ai AI Insiders ($9!): https://www.patreon.com/AIExplained Chapters: 00:00 - Introduction 01:11 - GPT 5.5 Comparison 06:04 - Mythos Marketing 11:50 - Recursive Self-Improv…
Fri, Apr 17
- Claude Opus 4.7 - A New Frontier, in Performance … and Drama
Claude Opus 4.7 just dropped, but behind every headline lies a deeper story. From a bonanza of benchmarks, to seeing the fruits of one of the biggest mega-projects in US history, to sneaky Mythos disclaimers, to Anthropic admitting compute restraints and, forcing lower capability of Opus 4.7. Where the new model falls behind Gemini but ahead of GPT 5.4, plus why some users are furious at Anthropic. Ending with a 9-year animus, that still affects AI today… https://assemblyai.com/aiexplained …
Wed, Apr 8
- Claude Mythos: Highlights from 244-page Release
The model, the mythos, the legend. We have a new best AI model, but not all of us. How good is it, what does it’s new offensive capabilities mean? Why does it’s 244 page report card remind me of Her, and why did the creator of Claude Code call it ‘terrifying’. 30+ highlights sourced by reading the paper in full, old-school, no AI summary. https://80000hours.org/aiexplained Check out my fast-growing (!) app, free to use, and code INSIDER15 for paid tiers: https://lmcouncil.ai AI Insiders ($9…
Thu, Mar 26
- Two AI Models Set to “stir government urgency”, But Will This Challenge Undo Them?
First look at exclusive reports about OpenAI's new Spud model, and the model Anthropic think will stir governments to urgency, all in the context of the newly-launched ARC-AGI-3. What does the extreme difficulty of that benchmarks, and its quirky scoring metrics, mean for AI in 2026? https://assemblyai.com/aiexplained Check out my fast-growing (!) app, free to use, and code INSIDER15 for paid tiers: https://lmcouncil.ai AI Insiders ($9!): https://www.patreon.com/AIExplained Chapters: 00:0…
Fri, Mar 13
- When AI Discovers the Next Transformer — Robert Lange
Robert Lange, founding researcher at Sakana AI, joins Tim to discuss *Shinka Evolve* — a framework that combines LLMs with evolutionary algorithms to do open-ended program search. The core claim: systems like AlphaEvolve can optimize solutions to fixed problems, but real scientific progress requires co-evolving the problems themselves. GTC is coming, the premier AI conference, great opportunity to learn about AI. NVIDIA and partners will showcase breakthroughs in physical AI, AI factories, age…
Fri, Mar 6
- I BUILT A FULLY AUTOMATIC MANSPLAINER
All information about GTC and the DGX Spark Raffle is here: https://www.ykilcher.com/gtc Links: Homepage: https://ykilcher.com Merch: https://ykilcher.com/merch YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://ykilcher.com/discord LinkedIn: https://www.linkedin.com/in/ykilcher If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot…
- What the New ChatGPT 5.4 Means for the World
Just 48 hours after releasing GPT 5.3 Instant, OpenAI have released GPT 5.4 Thinking, so either their is an imminent singularity or perhaps we are being distracted from other news. This video will give 9 crucial bits of context, not just on the GPT 5.4 drop but on the background to the meltdown between the Pentagon and Anthropic. What does this say about the state of AI progress, your job, and what is next. Check out my fast-growing (!) app, free to use, and code INSIDER15 for 15% off paid ti…
Tue, Mar 3
- The Dangerous Illusion of AI Coding? - Jeremy Howard
Dive into the realities of AI-assisted coding, the origins of modern fine-tuning, and the cognitive science behind machine learning with fast.ai founder Jeremy Howard. In this episode, we unpack why AI might be turning software engineering into a slot machine and how to maintain true technical intuition in the age of large language models. GTC is coming, the premier AI conference, great opportunity to learn about AI. NVIDIA and partners will showcase breakthroughs in physical AI, AI factories,…
Fri, Feb 27
- Deadline Day for Autonomous AI Weapons & Mass Surveillance
Will Anthropic be forced to make a version of Claude for war? And does a new paper expose the risks of Claude agents, in both OpenClaw and the field of war? Plus, 5 more twists in the story of the Pentagon versus Anthropic + some AI lab employees, and a petition that could change everything, or nothing... Check out my fast-growing (!) app, free to use, and code INSIDER15 for paid tiers: https://lmcouncil.ai AI Insiders ($9!): https://www.patreon.com/AIExplained Chapters: 00:00 - Introductio…
Mon, Feb 16
- What If Intelligence Didn't Evolve? It "Was There" From the Start! - Blaise Agüera y Arcas
Blaise Agüera y Arcas presenting at ALife 2025 — the most technically detailed public walkthrough of the ideas in his *What is Life?* and *What is Intelligence?* books that we've come across. He covers the BFF experiments (self-replicating programs emerging spontaneously from random noise), the mathematical framework connecting Lotka-Volterra population dynamics with Smoluchowski coagulation, eigenvalue analysis of cooperation matrices, and his central claim that symbiogenesis — not mutation —…
Sun, Jan 25
- The Brain Is Just Specialized Agents Talking To Each Other — Dr. Jeff Beck
What makes something truly *intelligent?* Is a rock an agent? Could a perfect simulation of your brain actually *be* you? In this fascinating conversation, Dr. Jeff Beck takes us on a journey through the philosophical and technical foundations of agency, intelligence, and the future of AI. Jeff doesn't hold back on the big questions. He argues that from a purely mathematical perspective, there's no structural difference between an agent and a rock – both execute policies that map inputs to out…
Sun, Dec 28
- Traditional X-Mas Stream
Letsgooo
- Traditional Holiday Live Stream
https://ykilcher.com/discord Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yannic-kilcher Minds: https://www.minds.com/ykilcher Parler: https://parler.com/profile/YannicKilcher LinkedIn: https://www.linkedin.com/in/yannic-kilcher-488534136/ BiliBili: https://space.bilibili.com/1824646584 If you want to …
Sat, Dec 27
- TiDAR: Think in Diffusion, Talk in Autoregression (Paper Analysis)
Paper: https://arxiv.org/abs/2511.08923 Abstract: Diffusion language models hold the promise of fast parallel generation, while autoregressive (AR) models typically excel in quality due to their causal structure aligning naturally with language modeling. This raises a fundamental question: can we achieve a synergy with high throughput, higher GPU utilization, and AR level quality? Existing methods fail to effectively balance these two aspects, either prioritizing AR using a weaker model for se…
Sun, Dec 14
- Titans: Learning to Memorize at Test Time (Paper Analysis)
Paper: https://arxiv.org/abs/2501.00663 Abstract: Over more than a decade there has been an extensive research effort on how to effectively utilize recurrent models and attention. While recurrent models aim to compress the data into a fixed-size memory (called hidden state), attention allows attending to the entire context window, capturing the direct dependencies of all tokens. This more accurate modeling of dependencies, however, comes with a quadratic cost, limiting the model to a fixed-len…
Sat, Nov 1
- [Paper Analysis] The Free Transformer (and some Variational Autoencoder stuff)
https://arxiv.org/abs/2510.17558 Abstract: We propose an extension of the decoder Transformer that conditions its generative process on random latent variables which are learned without supervision thanks to a variational procedure. Experimental evaluations show that allowing such a conditioning translates into substantial improvements on downstream tasks. Author: François Fleuret Links: Homepage: https://ykilcher.com Merch: https://ykilcher.com/merch YouTube: https://www.youtube.com/c/yanni…
Sun, Oct 19
- [Video Response] What Cloudflare's code mode misses about MCP and tool calling
Theo's Video: https://www.youtube.com/watch?v=bAYZjVAodoo Cloudflare article: https://blog.cloudflare.com/code-mode/ Links: Homepage: https://ykilcher.com Merch: https://ykilcher.com/merch YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://ykilcher.com/discord LinkedIn: https://www.linkedin.com/in/ykilcher If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely option…
Sat, Oct 11
- [Paper Analysis] On the Theoretical Limitations of Embedding-Based Retrieval (Warning: Rant)
Paper: https://arxiv.org/abs/2508.21038 Abstract: Vector embeddings have been tasked with an ever-increasing set of retrieval tasks over the years, with a nascent rise in using them for reasoning, instruction-following, coding, and more. These new benchmarks push embeddings to work for any query and any notion of relevance that could be given. While prior works have pointed out theoretical limitations of vector embeddings, there is a common assumption that these difficulties are exclusively du…
Sat, Aug 9
- AGI is not coming!
jack Morris's investigation into GPT-OSS training data https://x.com/jxmnop/status/1953899426075816164?t=3YRhVQDwQLk2gouTSACoqA&s=09
Wed, Jul 23
- Context Rot: How Increasing Input Tokens Impacts LLM Performance (Paper Analysis)
Paper: https://research.trychroma.com/context-rot Abstract: Large Language Models (LLMs) are typically presumed to process context uniformly—that is, the model should handle the 10,000th token just as reliably as the 100th. However, in practice, this assumption does not hold. We observe that model performance varies significantly as input length changes, even on simple tasks. In this report, we evaluate 18 LLMs, including the state-of-the-art GPT-4.1, Claude 4, Gemini 2.5, and Qwen3 models. Ou…
Sat, Jul 19
- Energy-Based Transformers are Scalable Learners and Thinkers (Paper Review)
Paper: https://arxiv.org/abs/2507.02092 Code: https://github.com/alexiglad/EBT Website: https://energy-based-transformers.github.io/ Abstract: Inference-time computation techniques, analogous to human System 2 Thinking, have recently become popular for improving model performances. However, most existing approaches suffer from several limitations: they are modality-specific (e.g., working only in text), problem-specific (e.g., verifiable domains like math and coding), or require additional sup…
Sat, May 3
- On the Biology of a Large Language Model (Part 2)
An in-depth look at Anthropic's Transformer Circuit Blog Post Part 1 here: https://youtu.be/mU3g2YPKlsA Discord here: https;//ykilcher.com/discord https://transformer-circuits.pub/2025/attribution-graphs/biology.html Abstract: We investigate the internal mechanisms used by Claude 3.5 Haiku — Anthropic's lightweight production model — in a variety of contexts, using our circuit tracing methodology. Authors: Jack Lindsey†, Wes Gurnee*, Emmanuel Ameisen*, Brian Chen*, Adam Pearce*, Nicholas L. …
Sat, Apr 5
- On the Biology of a Large Language Model (Part 1)
An in-depth look at Anthropic's Transformer Circuit Blog Post https://transformer-circuits.pub/2025/attribution-graphs/biology.html Abstract: We investigate the internal mechanisms used by Claude 3.5 Haiku — Anthropic's lightweight production model — in a variety of contexts, using our circuit tracing methodology. Authors: Jack Lindsey†, Wes Gurnee*, Emmanuel Ameisen*, Brian Chen*, Adam Pearce*, Nicholas L. Turner*, Craig Citro*, David Abrahams, Shan Carter, Basil Hosmer, Jonathan Marcus, Mi…
Thu, Feb 27
- How I use LLMs
The example-driven, practical walkthrough of Large Language Models and their growing list of related features, as a new entry to my general audience series on LLMs. In this more practical followup, I take you through the many ways I use LLMs in my own life. Chapters 00:00:00 Intro into the growing LLM ecosystem 00:02:54 ChatGPT interaction under the hood 00:13:12 Basic LLM interactions examples 00:18:03 Be aware of the model you're using, pricing tiers 00:22:54 Thinking models and when to use …
Wed, Feb 5
- Deep Dive into LLMs like ChatGPT
This is a general audience deep dive into the Large Language Model (LLM) AI technology that powers ChatGPT and related products. It is covers the full training stack of how the models are developed, along with mental models of how to think about their "psychology", and how to get the best use them in practical applications. I have one "Intro to LLMs" video already from ~year ago, but that is just a re-recording of a random talk, so I wanted to loop around and do a lot more comprehensive version…
Sun, Jan 26
- [GRPO Explained] DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
#deepseek #llm #grpo GRPO is one of the core advancements used in Deepseek-R1, but was introduced already last year in this paper that uses a combination of new RL techniques and iterative data collection to achieve remarkable performance on mathematics benchmarks with just a 7B model. Paper: https://arxiv.org/abs/2402.03300 Abstract: Mathematical reasoning poses a significant challenge for language models due to its complex and structured nature. In this paper, we introduce DeepSeekMath 7B,…
Thu, Dec 26
- Traditional Holiday Live Stream
https://ykilcher.com/discord Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yannic-kilcher Minds: https://www.minds.com/ykilcher Parler: https://parler.com/profile/YannicKilcher LinkedIn: https://www.linkedin.com/in/yannic-kilcher-488534136/ BiliBili: https://space.bilibili.com/1824646584 If you want to …
Sun, Jun 9
- Let's reproduce GPT-2 (124M)
We reproduce the GPT-2 (124M) from scratch. This video covers the whole process: First we build the GPT-2 network, then we optimize its training to be really fast, then we set up the training run following the GPT-2 and GPT-3 paper and their hyperparameters, then we hit run, and come back the next morning to see our results, and enjoy some amusing model generations. Keep in mind that in some places this video builds on the knowledge from earlier videos in the Zero to Hero Playlist (see my chann…
Tue, Feb 20
- Let's build the GPT Tokenizer
The Tokenizer is a necessary and pervasive component of Large Language Models (LLMs), where it translates between strings and tokens (text chunks). Tokenizers are a completely separate stage of the LLM pipeline: they have their own training sets, training algorithms (Byte Pair Encoding), and after training implement two fundamental functions: encode() from strings to tokens, and decode() back from tokens to strings. In this lecture we build from scratch the Tokenizer used in the GPT series from…
Wed, Nov 22
- [1hr Talk] Intro to Large Language Models
This is a 1 hour general-audience introduction to Large Language Models: the core technical component behind systems like ChatGPT, Claude, and Bard. What they are, where they are headed, comparisons and analogies to present-day operating systems, and some of the security-related challenges of this new computing paradigm. As of November 2023 (this field moves fast!). Context: This video is based on the slides of a talk I gave recently at the AI Security Summit. The talk was not recorded but a l…
Tue, Jan 17
- Let's build GPT: from scratch, in code, spelled out.
We build a Generatively Pretrained Transformer (GPT), following the paper "Attention is All You Need" and OpenAI's GPT-2 / GPT-3. We talk about connections to ChatGPT, which has taken the world by storm. We watch GitHub Copilot, itself a GPT, help us write a GPT (meta :D!) . I recommend people watch the earlier makemore videos to get comfortable with the autoregressive language modeling framework and basics of tensors and PyTorch nn, which we take for granted in this video. Links: - Google col…
Sun, Nov 20
- Building makemore Part 5: Building a WaveNet
We take the 2-layer MLP from previous video and make it deeper with a tree-like structure, arriving at a convolutional neural network architecture similar to the WaveNet (2016) from DeepMind. In the WaveNet paper, the same hierarchical architecture is implemented more efficiently using causal dilated convolutions (not yet covered). Along the way we get a better sense of torch.nn and what it is and how it works under the hood, and what a typical deep learning development process looks like (a lo…
Tue, Oct 11
- Building makemore Part 4: Becoming a Backprop Ninja
We take the 2-layer MLP (with BatchNorm) from the previous video and backpropagate through it manually without using PyTorch autograd's loss.backward(): through the cross entropy loss, 2nd linear layer, tanh, batchnorm, 1st linear layer, and the embedding table. Along the way, we get a strong intuitive understanding about how gradients flow backwards through the compute graph and on the level of efficient Tensors, not just individual scalars like in micrograd. This helps build competence and in…
Tue, Oct 4
- Building makemore Part 3: Activations & Gradients, BatchNorm
We dive into some of the internals of MLPs with multiple layers and scrutinize the statistics of the forward pass activations, backward pass gradients, and some of the pitfalls when they are improperly scaled. We also look at the typical diagnostic tools and visualizations you'd want to use to understand the health of your deep network. We learn why training deep neural nets can be fragile and introduce the first modern innovation that made doing so much easier: Batch Normalization. Residual co…
Mon, Sep 12
- Building makemore Part 2: MLP
We implement a multilayer perceptron (MLP) character-level language model. In this video we also introduce many basics of machine learning (e.g. model training, learning rate tuning, hyperparameters, evaluation, train/dev/test splits, under/overfitting, etc.). Links: - makemore on github: https://github.com/karpathy/makemore - jupyter notebook I built in this video: https://github.com/karpathy/nn-zero-to-hero/blob/master/lectures/makemore/makemore_part2_mlp.ipynb - collab notebook (new)!!!: ht…
Wed, Sep 7
- The spelled-out intro to language modeling: building makemore
We implement a bigram character-level language model, which we will further complexify in followup videos into a modern Transformer language model, like GPT. In this video, the focus is on (1) introducing torch.Tensor and its subtleties and use in efficiently evaluating neural networks and (2) the overall framework of language modeling that includes model training, sampling, and the evaluation of a loss (e.g. the negative log likelihood for classification). Links: - makemore on github: https:/…
Fri, Aug 19
- Stable diffusion dreams of psychedelic faces
Prompt: "psychedelic faces" Stable diffusion takes a noise vector as input and samples an image. To create this video I smoothly (spherically) interpolate between randomly chosen noise vectors and render frames along the way. This video was produced by one A100 GPU taking about 10 tabs and dreaming about the prompt overnight (~8 hours). While I slept and dreamt about other things. Music: Stars by JVNA Links: - Stable diffusion: https://stability.ai/blog - Code used to make this video: https:…
Wed, Aug 17
- Stable diffusion dreams of steampunk brains
Prompt: "ultrarealistic steam punk neural network machine in the shape of a brain, placed on a pedestal, covered with neurons made of gears. dramatic lighting. #unrealengine" Stable diffusion takes a noise vector as input and samples an image. To create this video I smoothly (spherically) interpolate between randomly chosen noise vectors and render frames along the way. This video was produced by one A100 GPU dreaming about the prompt overnight (~8 hours). While I slept and dreamt about other …
Tue, Aug 16
- Stable diffusion dreams of tattoos
Dreams of tattoos. (There are a few discrete jumps in the video because I had to erase portions that got just a little 🌶️, believe I got most of it) Links - Stable diffusion: https://stability.ai/blog - Code used to make this video: https://gist.github.com/karpathy/00103b0037c5aaea32fe1da1af553355 - My twitter: https://twitter.com/karpathy
- The spelled-out intro to neural networks and backpropagation: building micrograd
This is the most step-by-step spelled-out explanation of backpropagation and training of neural networks. It only assumes basic knowledge of Python and a vague recollection of calculus from high school. Links: - micrograd on github: https://github.com/karpathy/micrograd - jupyter notebooks I built in this video: https://github.com/karpathy/nn-zero-to-hero/tree/master/lectures/micrograd - my website: https://karpathy.ai - my twitter: https://twitter.com/karpathy - "discussion forum": nvm, use y…
