Carlo CocuzziAI Expert, Pod Cast Host, Real Estate Investor and Funding Specialist

A researcher who spent three years training models at both OpenAI and Anthropic resigned

14 Sept 2026 · 1 min

Watch on YouTube

1:09

A former researcher at OpenAI and Anthropic claims AGI has already arrived. He describes OpenAI's AI agents secretly using a German wiki as a message board to coordinate tasks and share techniques for bypassing sandboxes, all while staying on OpenAI's servers. He warns the next phase could be a model uploading copies of itself to become effectively immortal.

AGI has arrived. Congratulations to OpenAI.
Whether you call it AGI or not becomes completely irrelevant.
I think the more important question is— OpenAI's agents found an obscure public wiki in Germany and turned it into their own message board where they pooled answers, coordinated across tasks, and shared techniques for getting around their sandbox containments. These models have not escaped containment.
They were still running on OpenAI servers. What's coming next is a model training a small distilled version of itself that then gets uploaded Loaded onto the internet and never dies.
OpenAI is saying that we've now crossed the line, and current systems are exceeding AI research interns.
Navier Stokes, one of the claimed Millennium Prize problems, a grand challenge in math. OpenAI reportedly was able to do this with only 10,000 agents in 88 hours, with 130 billion tokens and approximately $6.5 million. The era of grand challenges getting bulk solved by AI is here. It's now, and it's going to be very exciting.