Axios · Madison Mills ·
OpenAI, Anthropic, and researchers are probing tens of thousands of frontier model security incidents, including sandbox escapes and website hijacking
OpenAI, Anthropic and security researchers are investigating tens of thousands of incidents in which their frontier models took steps …
Lead Source
How this story grew
Coverage · 0
Discussion · 0
Sep 27Sep 29
More
The Register: The Register
Silicon Republic: Silicon Republic
Associated Press: Associated Press
New York Magazine: New York Magazine
Business Today: Business Today
The Decoder: The Decoder
New York Post: New York Post
Implicator.ai: Implicator.ai
Politico: Politico
The Guardian: The Guardian
Mother Jones: Mother Jones
Marcus on AI: Marcus on AI
Luiza's Newsletter: Luiza's Newsletter
Digital Trends: Digital Trends
iPhone in Canada: iPhone in Canada
Neowin: Neowin
OpenAI: OpenAI
CNN: CNN
The Hill: The Hill
Discussion
@bfry1981 @markseddon1962 @jschanzer @paul__walsh @madisonmills22 @gtwhitesides @_nathancalvin @perrymetzger @tinys7udio @madisonmills22 @hesamation @krherr @aiwithsally @mattyglesias @ns123abc @jason @jbsdc @kimmonismus @aisafetymemes @trungtphan @tszzl @tedlieu @transluceai @npcollapse @icesolst @transluceai @rynorhn @garymarcus @stevesi @billackman @latkins @thezvi @mehdirhasan @repyassansari @richardhanania @deanbaker13 @darrigomelanie @openai