Anadolu | Anadolu | Getty Images
Jan Leike, one of OpenAI’s chief security researchers, who resigned from the artificial intelligence company earlier this month, said on Tuesday that he has joined rival artificial intelligence startup Anthropic.
Leike announced his resignation from OpenAI on May 15, days after the company disbanded the Super Alignment group he co-led. Founded in 2023, the team focuses on long-term artificial intelligence risks. OpenAI co-founder Ilya Sutskever announced his resignation in a post on X on May 14.
“I’m excited to join @AnthropicAI to continue the super-alignment mission,” Leike write on X. “My new team will work on scalable supervision, weak-to-strong generalization, and automatic alignment research.”
human support Amazonwhich has committed up to $4 billion for a minority stake in the company.
In an article after leaving OpenAI, Leike wrote: “Leaving this job was one of the hardest things I’ve ever done because we desperately need to figure out how to guide and control AI systems that are much smarter than us. “
Since OpenAI launched ChatGPT in late 2022, AI security has quickly become important across the tech landscape, triggering a boom in generative AI products and investments. Some in the industry have expressed concern that companies are releasing powerful artificial intelligence products to the public too quickly without fully considering the potential social harm.
OpenAI, supported by MicrosoftOn Tuesday it said it had established a new safety and security committee to be led by senior executives including Chief Executive Officer Sam Altman. The committee will recommend “security decisions for OpenAI projects and operations” to the company’s board of directors.
Anthropic was founded in 2021 by brother and sister Dario Amodei and Daniela Amodei, as well as other former OpenAI executives, and launched ChatGPT competitor Claude 3 in March. In addition to Amazon, the company also received support from Google, sales force and skyrocketing.
watch: Anthropic co-founder discusses artificial intelligence adoption