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.

We’re releasing a broad range of new mathematical results produced by an internal frontier model. We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and public recommendations to inform how we release these results.

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 the entire session.

Was testing out dots over the weekend and it figured out how to save me ~$500/yr in recurring charges. Most impressive part was when it connected to customer service and handled a whole texting convo on my behalf.

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.

A few thoughts on this: 1) If you’ve only seen clips of this interview, I’d encourage you to watch the full podcast. I push back on plenty of AI hype in it. 2) As I said in the podcast, this example is academic. My intention was to illustrate how hard it is to make absolute guarantees about isolation, which is why it's important to have layers of defense. The part before the clip starts is me talking about other layers of defense. 3) The example I'm bringing up isn't about weight exfiltration via temperature sensors, it's about coordination between agents that are supposed to be fully isolated and independent. Coordination can require very few bits of information. 4) One lesson from the HF incident is that we put too much trust in sandbox isolation and didn't have enough independent safeguards. Airgapping is an extremely strong safeguard. When designing safety protocols, I think it's much better to overestimate rather than underestimate.

Also, my team is hiring! We research long-horizon agents and multi-agent. We’re hiring for alignment/safety because we want to develop new research with alignment/safety in mind during the whole process. We’re also hiring for human-AI interaction.

Happy to finally do a deep dive on multi-agent with @dwarkesh_sp! None of it would have happened without the great work on multi-agent from my @OpenAI teammates @kevinleestone, @mikegmalek, @__eknight__, @amuellerml, @zhangir_azerbay, @CheukHeiChu, and many others.

@deredleritt3r It’s not a secret. It’s a combination of the HF incident, the capabilities of this new model, the concerning trajectory of monitorability, and the speed of improvement in capabilities.

The Navier Stokes solution was the result of a collaboration of ~10,000 (!) agents working together. Over the past year, we’ve been training models to collaborate through multiagent RL. It’s been amazing to see how much better the models have become at this: it seems clear now that one of the most effective ways to solve hard problems is to give models huge amounts of unstructured parallel test-time compute and let them decide how to work together.

This past week at OpenAI has been the most humbling experience of my life. We all feel the weight of what’s coming. When I joined OpenAI earlier this year, I speculated that AI might write an Annals paper in 2027—and maybe, just maybe, solve a Millennium Prize Problem in 2028. That was considered very bullish at the time. The first happened in May; the second, this September. What a surreal time to be alive!