【Music Archives】

OpenAI 's internal safety committee is Music Archivesbecoming an independent oversight body, as CEO Sam Altman leaves the group's top leadership role.

The announcement came as part of the results of a 90-day evaluation by the existing Altman-chaired Safety and Security Committee into the company's AI safeguards and governance.

The committee's new leadership includes Zico Kolter (Carnegie Mellon University), as well as existing members Adam D’Angelo (Quora) and Nicole Seligman (formerly Sony). "The Safety and Security Committee will be briefed by company leadership on safety evaluations for major model releases, and will, along with the full board, exercise oversight over model launches, including having the authority to delay a release until safety concerns are addressed," the announcement reads.

In May, the AI company launched a revamped safety and security team with Altman in the top seat, following the dissolution of its former security body and the loss of more than half of its relevant employees. Multiple former employees have raised concerns about Altman's leadership and OpenAI's safety protocols.

• ChatGPT was messaging users first — but OpenAI said this wasn’t supposed to happen
• Meet o1: OpenAI's advanced reasoning 'Strawberry' model
• OpenAI, Anthropic agree to have their models tested before making them public
• Apple is reportedly investing in OpenAI — Microsoft’s rival wants in on the ChatGPT maker, too
• OpenAI supports California's AI-watermarking bill

Last week, Altman told OpenAI employees that the company plans to overhaul its non-profit corporate structure in favor of a for-profit push. The company, the CEO said, had "outgrown" its existence as a capped for-profit LLC overseen by a non-profit board, but the shift was also reportedly an essential element to the AI startup's anticipated $150 billion valuation.

(Editor: {typename type="name"/})