will depue(@willdepue):openai people keep gesticulating towards the ‘new model‘ so much so that i’m absolutely sure it’s a new pretrain and not some RL run, which implies (i think) OpenAI has a model that not more than 2 weeks into pretraining was better than Astra and solved Navier-Stokes. insane

will depue

@willdepue

dei ex machina prev @ openai (sora 1 & 2, posttraining o3/4o, pretraining moonshots)

Joined May 2018

2.6K Following 77.4K Followers

will depue@willdepue

2026.09.15 06:25

openai people keep gesticulating towards the ‘new model‘ so much so that i’m absolutely sure it’s a new pretrain and not some RL run, which implies (i think) OpenAI has a model that not more than 2 weeks into pretraining was better than Astra and solved Navier-Stokes. insane

Noam Brown@polynoamial

2026.09.15 05:05

@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.