Nearly half of OpenAI's Safety Researchers have left

Nearly half of OpenAI's AGI safety team has departed in recent months, according to former researcher Daniel Kokotajlo. This exodus includes notable figures such as cofounder John Schulman and chief scientist Ilya Sutskever, who resigned amid concerns that safety considerations were being overshadowed by product development. Kokotajlo, who left OpenAI in April 2023, noted that the team, which once comprised about 30 members focused on mitigating the risks of artificial general intelligence (AGI), has dwindled to around 16. He expressed concerns about a "chilling effect" on research related to AGI safety, suggesting that the company's shift towards commercial interests has marginalized safety-focused researchers. Critics of the AI safety movement argue that the perceived threats of AGI are exaggerated and that the focus should instead be on leveraging AI to address pressing global issues. The recent departures raise questions about OpenAI's commitment to ensuring the safe development of AGI, especially as the company has opposed legislative efforts aimed at regulating AI development. Kokotajlo's observations reflect a broader concern about the influence of profit motives on safety considerations within the AI industry.

- Nearly half of OpenAI's AGI safety team has left the company.

- Departures include key figures who voiced concerns over prioritizing product development over safety.

- Critics argue that the focus on AGI risks may hinder innovation and overlook pressing global challenges.

- The exodus raises questions about OpenAI's commitment to AGI safety amid commercial pressures.

- Kokotajlo warns of a "chilling effect" on research related to AGI safety within the company.