OpenAI former safety researcher Jan Leike joins rival AI startup Anthropic

Despite its success in the generative AI space, OpenAI continues to experience the departure of several key members from its leadership team. The latest is Jan Leike, one of the company’s lead safety researchers who resigned earlier this month to join rival AI startup Anthropic.

In a post on X, Leike announced that he has joined Anthropic to “continue the superalignment mission.” Leike revealed his resignation from OpenAI on May 15, just days before the company disbanded the superalignment group he co-led. That team was originally formed in 2023 to address long-term AI risks.

“I’m excited to join @AnthropicAI to continue the superalignment mission. My new team will work on scalable oversight, weak-to-strong generalization, and automated alignment research.”

I'm excited to join @AnthropicAI to continue the superalignment mission!

My new team will work on scalable oversight, weak-to-strong generalization, and automated alignment research.

If you're interested in joining, my dms are open.

— Jan Leike (@janleike) May 28, 2024

“Stepping away from this job has been one of the hardest things I have ever done, because we urgently need to figure out how to steer and control AI systems much smarter than us,” Leike wrote in another post following his departure from OpenAI.

The departure of Leike comes a few days after OpenAI chief scientist and ChatGPT-4 creator Ilya Sutskever announced his departure from the company he co-founded nine years ago.

AI safety has rapidly become a critical focus in the tech sector since OpenAI’s launch of ChatGPT in late 2022, which triggered a surge in generative AI products and investments. However, some industry experts are concerned that companies are rushing to release these powerful AI technologies without fully considering their potential societal impacts.

Meanwhile, OpenAI on Tuesday announced the creation of a new safety and security committee led by senior executives, including CEO Sam Altman. The committee is tasked with recommending “safety and security decisions for OpenAI projects and operations” to the company’s board.

Founded in 2021 by Dario Amodei, former VP of research at OpenAI, alongside Jack Clark, Sam McCandlish, and Tom Brown, Anthropic aims to harness the potential of generative artificial intelligence for significant global impact. Sensing the transformative power of large-scale AI systems, Dario, along with his sister Daniela, embarked on a mission to create AI systems that are not only scalable but also interpretable and resilient. Before co-founding Anthropic, Daniela Amodei served as OpenAI’s vice president of safety and policy.

Jan Leike (Credit: CNN)