AIs Are Getting Increasingly Dangerous (opens in new tab)
Despite talk about an "option" to pause, top AI companies continue to race to superintelligence, even as their own tests show growing cyber, biosecurity, and loss-of-control risks.
Read the original article