David Robinson, a pacesetter on OpenAI’s Security Programs group, has resigned from the corporate.
An OpenAI spokesperson advised Enterprise Insider on Friday that Robinson left final week, marking yet one more second of upheaval within the firm’s security ranks. Robinson had beforehand led the corporate’s coverage planning.
Robinson labored on security transparency on the firm, together with serving to to develop and share system playing cards that present details about its fashions.
He didn’t instantly reply to requests for remark.
Robinson’s departure comes alongside a wave of consideration on OpenAI’s security practices and staffing. The corporate shared on Thursday that it had “parted methods” with three researchers, saying that they had violated its insurance policies for sharing delicate data.
Within the aftermath of the high-profile Hugging Face security breach, OpenAI is grappling with a sequence of different security incidents by which it mentioned its AI fashions have misbehaved.
Progress in mannequin capabilities has led researchers and firm executives to name for a slowdown in AI growth. OpenAI CEO Sam Altman, on the firm’s developer convention on Tuesday, mentioned security and alignment want to remain forward of mannequin capabilities.
Johannes Heidecke, the corporate’s security head, additionally left earlier this yr.
Robinson, in early September, wrote on X that he felt folks at OpenAI had been waking as much as the implications of extremely succesful AI fashions and likewise expressed some doubt.
Issues at OpenAI “are certainly altering considerably by the day, however I have no idea whether or not we’re altering quick sufficient,” he wrote on X.
Have a tip? Contact this reporter through electronic mail at scouncil@businessinsider.com, or over textual content, Sign, Telegram, or WhatsApp at 415-757-8198. Use a private electronic mail tackle, a nonwork WiFi community, and a nonwork machine; here’s our guide to sharing data securely.
