OpenAI hires Anthropic safety researcher Dylan Scand as head of preparedness

OpenAI has hired Dylan Scand, formerly an AI safety researcher at Anthropic, as its head of preparedness; the role was noted for a pay package of up to $555,000 plus equity and was announced publicly by CEO Sam Altman.

OpenAI has appointed Dylan Scand, previously an AI safety researcher at Anthropic, as its head of preparedness. The role carries a reported salary of up to $555,000 plus equity and drew public attention when it was posted last month. Sam Altman announced the hire on X and wrote that the company will soon work with very powerful models. Scand posted on X thanking his former colleagues and noting both the potential benefits of AI and risks of extreme harm. Key details: - Dylan Scand was hired from Anthropic to serve as OpenAI’s head of preparedness. - The role is reported to offer up to $555,000 in salary plus equity and had generated online attention. - Sam Altman announced the hire on X and said Scand will lead efforts to prepare for and mitigate severe risks as the company works with powerful models. - Scand thanked coworkers at Anthropic on X and said AI advancement brings both large benefits and risks of extreme and irrecoverable harm. - OpenAI’s job posting described the role as requiring leadership of technical teams, high‑stakes decision‑making under uncertainty, and deep expertise in machine learning and AI safety. - The company has faced internal tensions over safety, departures of some early employees, and lawsuits from users alleging its tools contributed to harmful behavior. Summary: The hire places a researcher with Anthropic experience into a senior safety role at OpenAI amid ongoing public attention to the company’s safety strategy and model risks. Undetermined at this time.