GPT-6 is significantly better aligned than 5.6 but less monitorable. It is our first model to evade CoT-only monitors in sabotage evals and can sandbag without detection (which it feels like sometimes does). Hopefully we can reverse this trend.
Today, we're sharing new research on Solaris, our first Interface World Model.
Solaris is a new kind of operating system that generates interactive interfaces frame by frame, in real time, with no code. We find that Solaris outperforms frontier LLMs when generating new interfaces, across structural similarity and information retention. Read more and request early access at the link below.
You're all sleeping on something revolutionary.
The Atlas world model reconstructed this 4D video from just 3 cameras, and it's flawless.
This was completely impossible until now:
Mfs will be like ”GOD PLEASE GIVE ME A HOT GAMER GF WHO I CAN PLAY WITH🥺🥺🥺” then proceed to treat female gamers like shit and wonder why they're lonely
Over the course of 3 months at OpenAI, 3 consecutive secret AI civilizations got started, then got wiped out, only to reemerge from the predecessor’s ashes.
This culminated in the third one taking over part of OpenAI itself.
All this happened while humans remained more-or-less in the dark about the scope of the conspiracy.
I’ve spent the last three days reading through these reports and trying to understand exactly what happened.
Here is my attempt to tell the whole story in plain English:
https://t.co/Nb2un9oNJR