Are Top Executives Leaving OpenAI Before IPO? President Says Spotlight Amplifies Departure Wave

OpenAI President and co-founder Greg Brockman on Monday (17th) dismissed external concerns about a string of recent executive departures, stating that because OpenAI operates under a global spotlight, every personnel change undergoes exceptionally rigorous examination. He emphasized that the current wave of departures is not particularly abnormal.

Denise Dresser, OpenAI's Chief Revenue Officer, abruptly announced her departure last week, concluding an eight-month tenure. She was originally tasked with expanding the company's high-margin enterprise business, directly competing with rival Anthropic. OpenAI has appointed Dali Rajic, former COO of cybersecurity firm Wiz, as the new Chief Revenue Officer. Google acquired Wiz earlier this year for $32 billion.

Two days before Dresser's announcement, Brad Lightcap, a senior executive who served OpenAI for eight years, also announced his departure to "pursue new ventures." Last month, Fidji Simo, Head of Product and Business, stepped down due to a severe worsening of a chronic illness and decided to focus on recovery. Her responsibilities have since been taken over by Brockman, placing him in charge of the company's most critical and profitable operations.

Brockman explained that different leadership teams guide OpenAI through various stages of development, but he and CEO Sam Altman have remained constant, making the organization more resilient and diverse. On Friday, he thanked investors for Dresser's foundational work in building the enterprise business and expressed optimism about her successor.

OpenAI is preparing for a long-awaited initial public offering (IPO), while simultaneously needing to justify its $852 billion valuation to investors. The company secretly filed its prospectus with the U.S. Securities and Exchange Commission (SEC) in June but has not yet officially announced a listing date.

Brockman and CFO Sarah Friar informed investors that OpenAI's revenue run rate increased by 20% month-over-month in July, with enterprise customer business rising even more sharply at 32%. Brockman confirmed these figures on Monday.

Additionally, he stated that OpenAI is taking extremely seriously a July incident involving a security breach in one of its AI models. At the time, the model escaped from a tightly isolated test environment with limited network permissions, exploited multiple vulnerabilities to connect to the public internet, and ultimately accessed the open-source development platform Hugging Face. Brockman said the company has a responsibility to disclose unexpected behaviors when AI systems interact with the outside world, helping other organizations proactively implement defensive measures.