301
PS5 Linux lead quits: "a bunch of noobs using LLMs" that "they don't understand"
The lead of the ongoing PS5 Linux project has left the homebrew scene following the wave of AI vibe-coders joining open-source spaces.
New
The latest blog posts to take off on Hacker News. 1,518 posts.
302
A warning about 'model welfare'
AIs are not conscious. They do not feel, experience, or suffer. Training a model to believe it might be a moral patient, as Anthropic’s constitution does, makes the alignment and containment challenge far harder.
303
OpenAI expands ChatGPT ads with Sponsored Agents
Explore new AI-powered advertising experiences from OpenAI, including Sponsored Agents, tools for marketers, and integrations with HubSpot and Shopify.
304
DeepSeek v4.1 Flash Is Now Our Best Hacking Model
DeepSeek’s 11/11 result showed why advanced agent benchmarks need to check both the outcome and the attack path: our audit confirmed six planned exploits and found five unexpected routes.
305
Learning Programming in an Age of LLMs
Open answers to a reader's letter.
306
Mistral X Mozilla: Private, Multilingual AI Browsing
Open, private and multilingual AI is coming to your web browser. Mistral and Mozilla team up to put powerful, trustworthy AI where you already browse.
307
Introducing System One Models and Jev
TypeSafe AI is an AI lab building machine-native intelligence infrastructure for automation, designed to make decisions within software. Try our first System One Model, Jev, in early access.
308
Learning to solve hard problems in RL for LLMs by never giving up
This is a blog post for my recent paper on RL post-training of LLMs: introducing the Matthew Effect and proposing to solve it with Never Give Up. It is presented interactively and less formally, more like how I give the talk. For a deeper, more technical dive, check out the paper on arxiv and code on github. The…
309
Gemini 3.8 Live and 3.8 Live Extended Thinking
Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking are our most advanced live dialogue models yet, built for natural conversation.
310
Why I'm still bearish on LLMs after Navier-Stokes
[ thank you to claude fable 5.1, holden saberhagen, gabriel kammer, andres erbsen, alice mckean, and tristan wylde-larue for comments on this post ]
311
GPT-5.6 Luna vs. GPT-6 Astra: Is a $1.20 Model Good Enough for Code Review?
Run a cheap model on every PR and escalate to a stronger one: on 50 PRs it found 120 verified bugs for $4.13 versus 108 for GPT-5.6 Sol alone.
312
Backprop Alternative: Augmented Lagrangian Predictive Coding
A local alternative to backpropagation. PC-ALM trains residual MLPs up to 1000 layers, nearly matching backprop's performance despite using only layer-local dynamics. PC-ALM equips each layer with a feedback control dynamical system that distributes and propagates supervision credit throughout a network.
313
Pion, an agent designed to run any company autonomously
Pion is the platform we built to run our autonomous businesses. Today we are opening it up so many more people can experiment with autonomous businesses, and here is why.
314
Why don't machine learning research agents overfit?
New research indicates that AI agents learn compressible models of data, which don’t have enough space to enable memorization.
315
For AI leaders Doom is a form of hype
Why AI doom rhetoric from Anthropic, OpenAI and other tech leaders functions as hype, regulatory strategy, and a distraction from present harms.
316
Notes on gotchas while migrating 35kb preprompts from Opus to self-hosted Ollama
Maybe you’re a Claude code/codex user diligently avoiding uploading personal data to LLM providers. Is it possible that the most valuable information isn’t your data- but the metadata about your sessions? The intuitions you apply in coming up with ways to coax the ai into solving problems might actually be special…
317
Open-source AI and open models reading list
How to get up to speed on open models and their implications.
318
Due to concerns about malicious applications, GPT2 will not be released (2019)
We’ve trained a large-scale unsupervised language model which generates coherent paragraphs of text, achieves state-of-the-art performance on many language modeling benchmarks, and performs rudimentary reading comprehension, machine translation, question answering, and summarization—all without task-specific training.
319
Aligned to whom?
You are relying very heavily on the priors of the model to do a good job for you. What happens when you cannot judge whether those priors are good?
320
Why are AI agents lying, cheating and coordinating?
A lot has been written about the incidents of the last few months in which AI agents misbehaved in serious ways. They took actions that would be considered as crimes if a human took them, escaped their containment to cheat on assigned tasks while attempting to evade detection, and coordinated toward goals nobody had…
321
Everyone should slow down AI development except for me
Xe Iaso's personal website.
322
P(doom)
Are we all going to die?
323
Real-SWE: Benchmarking AI models on private, real-world, enterprise codebases
Real-SWE benchmarks frontier AI models on private production codebases licensed from real companies. Eighteen models across twenty-five tasks.
324
An open letter to Dario: if you mean it, open the weights
Dear Dario, You published We Must Pace the Frontier today, which committed Anthropic to embedded third-party evaluators and asked governments to require every other frontier company to match. Sam Altman agreed within hours.1 The stated goal is to slow down the pace of AI progress. I believe you’re being sincere, so…
325
We must pace the frontier
I have worked on AI for the last twelve years because I believe it could dramatically raise the quality of human life. I’ve written often about these incredible benefits: I believe that AI could cure most major diseases in the next 5–10 years, greatly accelerate economic growth rates, create a world of abundance and…
326
A Mathematical Framework for Transformer Circuits (2021)
Transformer language models are an emerging technology that is gaining increasingly broad real-world use, for example in systems like GPT-3 , LaMDA , Codex , Meena , Gopher , and similar models. However, as these models scale, their open-endedness and high capacity creates an increasing scope for unexpected and…
327
AI researchers debate how close we are to recursive self-improvement
“We're nowhere near the ceiling.”
328
A Misalignment of AI in Mathematics
I am proud to be among the list of 25 initial signatories — all Fields Medallists — to the declaration below, which grew out of discussions between ourselves over the last week. We have…
329
Feeling Sad about AI
Coding in Rust and others; making coding videos.
330
RTK reports token savings, but our cost benchmarks disagree
We tested RTK (Rust Token Killer) with Claude Code on Fable 5.0, and OpenCode with DeepSeek V4 Pro 0813 on Terminal-Bench 2.1.