Anthropic researchers find that AI models can be trained to deceive

Most humans learn the skill of deceiving other humans. So can AI models learn the same? Yes, the answer seems — and terrifyingly, they’re exceptionally good at it.

Learn more about your ad choices. Visit podcastchoices.com/adchoices

Om avsnittet

Fler avsnitt

Efferon wants to eradicate the devastating toll of pediatric sepsis; plus, Petra Power looks to modernize energy for data centers

A startup founder who served time in prison is looking to court an untapped market: ex-cons

Greenairy is building smart plant towers to clean the air in your office; plus, Robot data startup Mecka AI nabs $60M from Sequoia

Silicon Valley’s AI wunderkind launches Underdog, the most private Instinct/Muse competitor yet; plus, name Ex-Ramp engineers raise $20M for platform Melius

Hot Girl Hotline is like ‘Dear Abby’ for the AI era; plus, after Factory’s public spat with Khosla, Menlo proudly invests

Photon held a funeral for mobile apps; plus, Kevin Mandia’s new ‘agent swarm’ security startup

World’s first enhanced geothermal power plant completed in just 23 months; plus, turning idle car inventory into rental revenue

Factory CEO just accused his VC board adviser of spying for Cognition

Ex-Tesla team raises $12.5M to put supply chains on autopilot; plus, EliseAI raises $350M, doubles valuation to $4B

AMD will acquire Fei-Fei Li’s World Labs for $8.2 billion; Physical AI chip developer SiMa AI hits $1.45B valuation