OpenAI Poaches Anthropic Safety Expert for Key Preparedness Role

OpenAI Recruits From Its Biggest Rival

OpenAI has turned to its primary competitor to fill one of the most closely watched positions in the AI industry. Dylan Scandinaro, who previously held an AGI safety role at Anthropic, has been appointed as OpenAI's new Head of Preparedness, commanding a base salary of up to five hundred and fifty-five thousand dollars annually plus equity.

CEO Sam Altman announced the hire on X, calling Scandinaro "by far the best candidate I have met, anywhere, for this role" and warning that "things are about to move quite fast" as the company works with increasingly powerful models.

A Role Vacant for Over a Year

The position has been empty since Aleksander Madry transitioned to a different role in July twenty twenty-four. OpenAI posted the job listing in December, with Altman cautioning prospective applicants that it would be "a stressful job" requiring them to "jump into the deep end pretty much immediately."

As Head of Preparedness, Scandinaro will oversee OpenAI's preparedness framework, which tracks frontier AI capabilities that could create new pathways to severe harm. The role involves building capability evaluations, threat models, and mitigations forming what OpenAI describes as a "coherent, rigorous, and operationally scalable safety pipeline."

Safety Concerns and Departures

The hire arrives against a backdrop of sustained criticism over OpenAI's safety culture. Former employees have accused the company of prioritising commercial products over safety research. Jan Leike, a former safety team co-lead, wrote upon departing that "safety culture and processes have taken a backseat to shiny products." Nearly half of the staff once focused on long-term AI risks had reportedly left the company by twenty twenty-four.

IPO Ambitions Add Urgency

The appointment also comes as OpenAI positions itself for a potential fourth-quarter twenty twenty-six initial public offering. Currently valued at approximately five hundred billion dollars, the company is holding informal discussions with investment banks and has expanded its finance team. OpenAI is racing to beat Anthropic to become the first major generative AI startup to go public, while simultaneously pursuing a funding round that could value the firm at eight hundred and thirty billion dollars.

Scandinaro acknowledged the gravity of the challenge ahead, writing that "the potential benefits are great, and so are the risks of extreme and even irrecoverable harm. There is a lot of work to do, and not much time to do it."