OpenAI is focusing on advancing its ChatGPT model on long-term research, reportedly leading to the departure of key personnel.
Shift to LLMs Spurs Staff Departures
The Sam Altman-led company’s strategy to allocate more resources to enhancing its large language models has led to the exit of several senior employees, including Vice President of Research Jerry Tworek, Model Policy Researcher Andrea Vallone, and Economist Tom Cunningham, The Financial Times reported on Tuesday.
OpenAI’s Chief Research Officer Mark Chen told the publication that the company remains committed to long-term foundational research and added that combining research with real-world deployment strengthens the company’s science by speeding up feedback, learning cycles, and rigor.
However, the Financial Times reported that OpenAI has increasingly prioritized large language models, with researchers on non-LLM projects seeing funding requests denied or scaled back. Teams behind Sora and DALL-E were reportedly under-resourced, while several non-language projects were wound down amid broader team reorganization.
OpenAI did not immediately respond to Benzinga‘s request for comment.
Slow Hiring Amid Expansion Challenges
OpenAI’s focus on ChatGPT follows Altman’s remarks that the company can “dramatically” slow hiring while still boosting output, as AI enables it to achieve more without rapidly expanding its workforce.
In December, OpenAI declared a “Code Red” crisis, pausing its monetization strategy to focus on product quality following the launch of Google‘s (NASDAQ:GOOGL) (NASDAQ:GOOG) Gemini 3. OpenAI’s revenue topped $20 billion in 2025, reflecting strong AI demand, but the company reportedly burns over $17 billion annually, with subscription income likely insufficient to cover its costly, compute-heavy operations. The company is also under pressure to demonstrate the revenue potential that justifies its $500 billion valuation.
In January, the company announced plans to test advertisements on the platform, free and Go tiers. The company has said that any ads would be clearly labeled, kept separate from ChatGPT responses, would not affect outputs, and user conversations would remain private, with paid tiers staying ad-free.
Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.
Image via Shutterstock
Recent Comments