OpenAI Safety Panel Led by Zico Kolter Gains Regulatory Power

What happened

Zico Kolter, a professor at Carnegie Mellon University, leads a four-person Safety and Security Committee at OpenAI. This panel has the authority to halt the release of new AI systems if they are deemed unsafe. The role gained increased importance after California and Delaware regulators made Kolter’s oversight a condition for OpenAI’s new business structure approval.

Why it matters

• The committee can prevent the release of AI technologies that pose safety or security risks.
• Kolter’s oversight is now a regulatory requirement, reflecting growing government scrutiny of AI safety.
• The panel’s decisions influence how OpenAI balances innovation with public safety in AI deployment.

Key facts

1. 1. The Safety and Security Committee at OpenAI is composed of four members and is chaired by Zico Kolter.
2. 2. Kolter was appointed chair of the committee more than a year ago, with his role gaining prominence in late 2025 due to regulatory agreements.
3. 3. OpenAI’s new business structure approval by California and Delaware regulators includes Kolter’s safety oversight as a key condition.