Another leader of OpenAI’s security strategy is leaving, joining a growing list of employees who have left the company’s security and alignment teams in recent years.

Johannes Heidecke, head of OpenAI’s Security Systems team, is leaving as the company reorganizes its research and security work under a single leader.

“We are grateful for Johannes’ contributions to OpenAI,” Mark Chen, director of research at OpenAI, told Business Insider. “We are excited about this next chapter under Mia Glaese’s leadership in research and security.”

OpenAI said it would integrate security more deeply into its research teams under Glaese, who will now serve as its vice president of research and security.

The shakeup comes as OpenAI reconsiders the relationship between its research and its security efforts. A spokesperson told Business Insider that “you can’t make good security decisions without understanding the underlying capabilities of the model, and you can’t make good research decisions without understanding the security implications.”

OpenAI’s mission is to ensure that artificial general intelligence — when it finally arrives — benefits all humanity. “Safety, the practice of enabling the positive impacts of AI while mitigating the negative ones, is therefore fundamental to our mission,” the company says.

OpenAI, however, has a poor track record of retaining its top security leaders, and some departing employees have publicly questioned its commitment to that job.

When Jan Leike, who once co-led OpenAI’s super-alignment team, left in May 2024, he said the company was prioritizing product launches over ensuring their security.

“In recent years, safety culture and processes have taken a backseat to shiny products,” Leike wrote in a post on X at the time. He said OpenAI needed to devote more resources to preparing future models and that building machines smarter than humans was an “inherently dangerous endeavor.”

Anthropic co-founder Dario Amodei was vice president of research at OpenAI before leaving in 2020 over disagreements over the company’s direction and approach to AI safety. A year later he founded Anthropic as a manufacturer of AI models committed to security.

Anthropic has suffered its own departures, most notably in February when Mrinank Sharm, who led its Safeguards Research Team, left and publicly warned that AI companies face pressure to “leave aside what matters most.” Dylan Scand, a security researcher at Anthropic, also left this year to become head of preparation at OpenAI.

Still, Anthropic’s core security leadership has remained largely intact compared to OpenAI. Andrew Curran, a prominent AI analyst, called OpenAI’s ever-revolving security door a curse in reference to Harry Potter.

Here are eight leaders of OpenAI’s security-focused teams who have left in recent years.

Ilya Sutskever, co-founder

Sutskever, OpenAI’s co-founder and its longtime chief scientist, left in May 2024. He had co-led the company’s Superalignment initiative with Leike, which was created to develop methods to control future superintelligent AI.

When he left, he said in a post on X that OpenAI’s progress had been “nothing short of miraculous.” He later founded Safe Superintelligence, an artificial intelligence laboratory dedicated exclusively to his namesake.

Jan Leike, co-founder

Leike left days after Sutskever. He offered one of the most direct public criticisms of OpenAI’s approach to security, saying it had reached a “breaking point” after protracted disagreements with leaders.

OpenAI disbanded the Superalignment team after Leike and Sutskever left and distributed its remaining employees among other research groups. Leike joined Anthropic to work on alignment research.

Miles Brundage, Senior Advisor for AGI Preparation

Brundage left OpenAI in October 2024 after six years.

“Neither OpenAI nor any other cutting-edge lab is ready, and the world is not ready either,” Brundage wrote in an essay on his Substack at the time, referring to the impact AI could have on the world.

He wrote that staying with OpenAI is an implicit agreement with the company’s values. “Anyone working at OpenAI should take seriously the fact that their actions and statements contribute to the organization’s culture and can create positive or negative dependencies as the organization begins to manage extremely advanced capabilities,” he wrote.

Lilian Weng, vice president of research and security

Weng, who previously led OpenAI’s Security Systems team, left the company in November 2024 after nearly seven years, according to his LinkedIn profile.

He did not publicly express his concerns about OpenAI in his exit statement. Instead, he praised the environment and said his work to “train these models to be powerful and responsible has set new standards in the industry.”

Andrea Vallone, Head of Model Policy Security Research

Vallone left at the end of 2025 after leading the team that shaped how OpenAI models respond in sensitive situations, including conversations involving emotional dependency and mental health crises.

His departure was initially announced internally and he did not publicly criticize OpenAI. Vallone later joined Anthropic’s alignment team to continue studying model behavior and safety.

Aleksander Madry, former head of preparation

Madry joined OpenAI to lead its Preparedness team, which assesses whether increasingly capable models could generate serious risks. He was removed from that leadership position in 2024 and focused his attention on AI reasoning before leaving the company in May 2026.

Joshua Achiam, former chief futurist

Achiam left in July. He previously led Mission Alignment, a group created to ensure the company’s AI development and deployment align with its nonprofit mission.

“There’s no specific reason for me to leave, or a specific reason why now. But it’s something I’ve been thinking about for a while and it feels right to me. The world now knows the secret and it seems possible to work on the mission from outside the walls of a border lab,” he wrote in a post on X in July.

Johannes Heidecke, head of security systems

Heidecke joined OpenAI in 2021 and took over Safety Systems following Weng’s departure. His team helped assess and mitigate risks in OpenAI models before their release.

Heidecke has not made any public statements since his planned departure was made public on Friday. OpenAI said Heidecke decided to leave because the company more closely integrated security work with its broader research organization.