Fired OpenAI Safety Researchers Speak Out + We Look Inside Pandora’s Box

OpenAI’s recent termination of safety researchers Mikita Balesni and Tomek Korbak highlights growing internal friction regarding AI safety and transparency. The researchers, who played central roles in coordinating independent audits of the Hugging Face incident, contend that their firing was a retaliatory measure for raising concerns about model monitorability and safety standards. This conflict underscores the broader challenge of maintaining rigorous, independent oversight within private AI labs driven by competitive and commercial pressures. Beyond the interview, the episode examines emerging tech industry trends, including the repurposing of dead malls into data centers, the complexities of corporate espionage, and the use of AI agents in gaming environments. These developments reflect the ongoing struggle to balance rapid technological innovation with the necessity for robust, external checks and balances in the development of frontier AI systems.