According to 1M AI News monitoring, Joanne Jang, Managing Director of OpenAI Labs, posted on X to announce her departure, ending her 4.5-year tenure at OpenAI. When she joined, the company was still a research lab of fewer than 200 people.
Jang contributed to the launches of GPT-4, DALL·E 2, text-to-speech, the Chat API, and memory capabilities, and helped build “Model Behavior” into an independent academic direction. She was responsible for the character design and post-training reinforcement learning for GPT-4o, GPT-4.5, and o3, and she also led the development of OpenAI’s AI behavior guidelines, the Model Spec. In September 2025, she transferred from the Model Behavior team to found OpenAI Labs, exploring a new interaction paradigm for collaboration between humans and AI. In 2025, Jang was named to TIME’s 100 Most Influential AI Figures in the “Thinkers” category.
In her farewell message, she said that if she could only be remembered for one thing, “I hope it’s that I brought frogs and pink text into OpenAI.”