Noam Brown is a researcher at OpenAI working on AI reasoning, games and safety. Recent coverage has centered on his posts about agent teams using over 100 billion tokens on the Navier-Stokes Millennium Prize problem and then applying a similar approach to red-teaming, LLMs surpassing top human experts on some research problems, the GPT-6 Sol and Luna releases with Luna priced at $0.10 input and $0.50 output per million tokens, and his responses to plagiarism accusations and questions about the prompts behind the math result.
Every winter, @MIT has run a 1-month class (6.9630) in which students make poker AIs and compete against each other for cash prizes. Coding agents have accelerated that process so much that it's now possible to build a poker bot in an *hour*. Also, I'll be at…
Only 2.5 years ago, AI labs would highlight their frontier LLM performance on grade school math benchmarks (GSM8K). I can't wait to see where we'll be in another 2.5 years. [Quoting @_AashishReddy]: It just never gets old
Last month a large team of agents used over 100 billion tokens to resolve the Navier-Stokes Millennium Prize problem. Now we're using a similar approach for red-teaming to find and fix infra vulnerabilities. [Quoting @Eric_Wallace_]: To make sure our frontier…
This phishing email made it past Gmail’s filters, but my dot caught it and warned me.
LLMs have crossed an important threshold: surpassing top human experts on some research problems. Capabilities have been steadily improving for years, but in scientific discovery there is a big difference between a model just below top human performance and o…
I met @ssokota 5 years ago at a conference poster session. I was impressed with his explanation, so I offered him an internship. We continued collaborating and he now works with me at @OpenAI. Later, he told me I was the only person to stop by his poster for…
I’m happy that @OpenAI presents model evals this way. We should measure intelligence as a function of cost. [Quoting @OpenAI]: GPT-6.1 Sol: near-Astra intelligence for a fifth of the price. It’s the most cost-efficient model for its performance available toda…
GPT-6 Sol and Luna are out, and they are better AND 50% cheaper than 5.6. Luna is now $0.10 input / $0.50 output per 1M tokens. This is on top of the 80% price cut to Luna we made at the end of July. Output went from $6 -> $0.50 within two months. [Quoting…
I met someone today whose landlord wanted to hike their rent by 10%. They asked ChatGPT, which pointed out that their unit was rent stabilized and an increase that large was illegal. They told their landlord, who backed off. This is my optimistic AI future. […
AI Safety Talk Slides Into Unbelievable ClaimsTwo viral AI safety claims this week showed how hard it is to separate AI fact from fiction. Andrew Yang said a lab head believed OpenAI's Hugging Face hacker bots planted self-replicating code across the internet; a security expert called that unlikely. OpenAI's Noam Brown said the incident showed people underestimated the AI and doubted air-gapping would stop it.
Very sad to see Levent double down on the plagiarism accusation. I hope my friends at @AnthropicAI stand up to this internally. It should be clear by now what the truth is.
I understand how, looking at today’s LLMs, people might think the only way we'd achieve NS is by using Levent’s/Tristan’s prompts. But I hope this plot conveys the model used is a huge step up from today’s LLMs. Nobody looked at Levent’s/Tristan’s prompts. Th…
Yes, this result cost millions of dollars. But remember that when @OpenAI announced o3 it cost ~$500,000 to score 87.5% on ARC-AGI 1. Today, Astra scores higher for ~$20. In 2025 it took us and GDM an enormous amount of compute to achieve IMO gold. For the 20…
The A.I. Daily