Anthropic’s New Safety Lead; The Startup Making Nvidia’s Software Easier To Use

We might be a long way away from sipping Pina Coladas on a beach while AI-powered humanoid robots handle all our work. But we’re already starting to see how AI is upending society. Some ChatGPT users are lamenting that they’ve lost their “only friend overnight” with the latest GPT-5 update, while other headlines suggest AI chatbots can even induce psychotic episodes. 

Partly in response to these trends, Anthropic has hired Dave Orr, who previously led Google’s efforts to incorporate large language models into its Google Assistant product, as its new head of safeguards to lead Anthropic’s safety work. He succeeds Vinay Rao, who left in April to become the chief technology officer for ROOST.tools, a nonprofit developing online safety tools. 

The new hire follows several developments in the last year which made safety researchers at Anthropic more concerned about the potential for AI to cause harm—whether intentionally or not.