Ex-Open AI researcher Jan Leike joins Anthropic amid AI safety concerns

Jan Leike, a prominent researcher — who recently resigned from OpenAI over safety and governance issues — has joined OpenAI competitor, Anthropic.

Leike announced his move on X, stating his new focus will be on “scalable oversight, weak-to-strong generalization, and automated alignment research.”

Leike’s departure from OpenAI was one of several recent high-profile exits based on the premise that “safety culture and processes have taken a backseat” at the ChatGPT creator.

This follows reported disagreements with OpenAI’s leadership over its direction, particularly concerns that the organization was prioritizing commercial interests over safety considerations in the race to develop advanced AI.

“We are long overdue in getting incredibly serious about the implications of AGI (artificial general intelligence). We must prioritize preparing for them as best as we can,” he said in a follow-up post on X post-resignation.

Leike’s resignation is not an isolated incident. Ilya Sutskever, OpenAI’s co-founder, also left earlier this month after reportedly being absent from the office since a failed coup attempt against CEO Sam Altman in November, last year.

Surskever has yet to announce his next move but his alignment with Anthropic’s values makes it a possible destination.

Leike’s background aligns with Anthropic’s mission. At OpenAI, he belonged to the “Superalignment” team focused on Ensuring AI systems remained aligned with human values. He criticized OpenAI for not allocating sufficient resources to guarantee the achievement of this goal.

“I believe much more of our bandwidth should be spent getting ready for the next generations of models, on security, monitoring, preparedness, safety, adversarial robustness, (super)alignment, confidentiality, societal impact, and related topics,” Leike said in a follow-up post elaborating his resignation. “These problems are quite hard to get right, and I am concerned we aren’t on a trajectory to get there.”

Anthropic, on the other hand, claims to be a responsible AI company and integrates ethical principles into AI development.

“Our research teams investigate the safety, inner workings, and societal impact of AI models — so that artificial intelligence has a positive impact on society as it becomes increasingly advanced and capable,” says the company’s mission statement.

OpenAI finally takes note

Following these high-profile exits, the OpenAI board appears to have taken notice. In a strategic move aligning with the direction advocated by Sutskever and Leike, the ChatGPT creator is establishing a “safety and security” board. This board will make recommendations on critical safety and security decisions on all OpenAI projects.

“The OpenAI Board formed a Safety and Security Committee led by directors Bret Taylor (Chair), Adam D’Angelo, Nicole Seligman, and Sam Altman (CEO). This committee will be responsible for making recommendations to the full Board on critical safety and security decisions for OpenAI projects and operations,” the company said in a statement.

The recent departures from OpenAI and the subsequent move to Anthropic highlight ongoing governance issues and ethical concerns within the rapidly evolving AI industry. As these developments unfold, the focus on safe and ethical AI continues to be a paramount concern for researchers and organizations alike.