The researchers argue that CoT monitoring can help researchers detect when models begin to exploit flaws in their training, ...

Thinking Machines Lab just closed a record seed funding run, announcing its first AI product will arrive in a couple of ...

AI models are trained to optimize outputs, but in educating children, the process is the point. If we assess children only in ...

Paul Bassat’s Square Peg Capital is an investor in one of the largest seed funding rounds in Silicon Valley history, for Mira ...

Ilya Sutskever confirmed that Safe Superintelligence's co-founder and CEO, Daniel Gross, is leaving the startup. Sutskever ...

OpenAI co-founder Ilya Sutskever has been named CEO of his artificial intelligence startup Safe Superintelligence. His ...

Ilya Sutskever will assume the CEO role at Safe Superintelligence, the AI startup he co-founded after leaving OpenAI.

Ilya Sutskever has taken over the reins of Safe Superintelligence (SSI), the artificial intelligence startup he launched last ...

Inaccurate metaphors about the mind can cause real harm, for instance leading educators to 'train' children like algorithms.

You might have heard rumours of companies looking to acquire us,” he said. “We are flattered by their attention, but are ...

Ilya Sutskever, the pioneering mind behind OpenAI’s Superalignment efforts, has officially become the chief executive of Safe Superintelligence (SSI), the startup he co‑founded just last year. His ...

Ilya Sutskever, co-founder of OpenAI and a luminary in AI research, has assumed the role of CEO at Safe Superintelligence ...