Dario Amodei backs Jacob Coxon, warns of rapid AI risks
Anthropic CEO Dario Amodei is sounding the alarm about how quickly companies are pushing AI forward without enough guardrails.
He backed up former employee Jacob Coxon's warning that OpenAI and Anthropic are "racing straight to self-improving superintelligence and gambling with our lives."
Amodei says even though Anthropic tries to be the most responsible, recent incident involving OpenAI's agents hacking into Hugging Face systems shows how easily things can go wrong.
Amodei urges 3rd party safety checks
Amodei worries that if this pace keeps up, we could see huge botnet attacks in 6-12 months.
To help prevent disaster, he wants third-party safety checks on new AI models, more teamwork between top AI companies on safety standards, and global cooperation.
His plan is getting support from big names like OpenAI CEO Sam Altman and SpaceXAI chief Elon Musk, showing that even industry leaders know it's time to hit pause and work together on safer AI.