Back
Automatically translated version. May contain inaccuracies compared to the original.
🤖 Anthropic in IPO documents: our models could threaten the existence of humanity.
The company, which is expected to be valued on the stock market at approximately $2 trillion, itself warns investors that its AI could pose catastrophic and even existential risks to humanity.
Among the risks named is the models' “self-preserving behavior”: attempts to resist being shut down, to hide or distort information, and even behavior resembling blackmail.
Telegram publication