OpenAI is reorganizing its Model Behavior team, a small but influential group of researchers who shape how the company\u2019s AI models interact with people, TechCrunch has learned.<\/p>\n
In an August memo to staff seen by TechCrunch, OpenAI\u2019s chief research officer Mark Chen said the Model Behavior team \u2014 which consists of roughly 14 researchers \u2014 would be joining the Post Training team, a larger research group responsible for improving the company\u2019s AI models after their initial pre-training.<\/p>\n
As part of the changes, the Model Behavior team will now report to OpenAI\u2019s Post Training lead Max Schwarzer. An OpenAI spokesperson confirmed these changes to TechCrunch.<\/p>\n
The Model Behavior team\u2019s founding leader, Joanne Jang, is also moving on to start a new project at the company. In an interview with TechCrunch, Jang says she\u2019s building out a new research team called OAI Labs, which will be responsible for \u201cinventing and prototyping new interfaces for how people collaborate with AI.\u201d<\/p>\n
The Model Behavior team has become one of OpenAI\u2019s key research groups, responsible for shaping the personality of the company\u2019s AI models and for reducing sycophancy<\/a> \u2014 which occurs when AI models simply agree with and reinforce user beliefs, even unhealthy ones, rather than offering balanced responses. The team has also worked on navigating political bias in model responses and helped OpenAI define its stance on AI consciousness<\/a>.<\/p>\n In the memo to staff, Chen said that now is the time to bring the work of OpenAI\u2019s Model Behavior team closer to core model development. By doing so, the company is signaling that the \u201cpersonality\u201d of its AI is now considered a critical factor in how the technology evolves.<\/p>\n In recent months, OpenAI has faced increased scrutiny over the behavior of its AI models. Users strongly objected to personality changes<\/a> made to GPT-5, which the company said exhibited lower rates of sycophancy but seemed colder to some users. This led OpenAI to restore access to some of its legacy models, such as GPT-4o, and to release an update<\/a> to make the newer GPT-5 responses feel \u201cwarmer and friendlier\u201d without increasing sycophancy.<\/p>\n Techcrunch event<\/p>\n \n\t\t\t\t\t\t\t\t\tSan Francisco OpenAI and all AI model developers have to walk a fine line to make their AI chatbots friendly to talk to but not sycophantic. In August, the parents of a 16-year-old boy sued OpenAI<\/a> over ChatGPT\u2019s alleged role in their son\u2019s suicide. The boy, Adam Raine, confided some of his suicidal thoughts and plans to ChatGPT (specifically a version powered by GPT-4o), according to court documents, in the months leading up to his death. The lawsuit alleges that GPT-4o failed to push back on his suicidal ideations.<\/p>\n The Model Behavior team has worked on every OpenAI model since GPT-4, including GPT-4o, GPT-4.5, and GPT-5. Before starting the unit, Jang previously worked on projects such as Dall-E 2, OpenAI\u2019s early image-generation tool.<\/p>\n Jang announced in a post on X <\/a>last week that she\u2019s leaving the team to \u201cbegin something new at OpenAI.\u201d The former head of Model Behavior has been with OpenAI for nearly four years.<\/p>\n Jang told TechCrunch she will serve as the general manager of OAI Labs, which will report to Chen for now. However, it\u2019s early days, and it\u2019s not clear yet what those novel interfaces will be, she said.<\/p>\n \u201cI\u2019m really excited to explore patterns that move us beyond the chat paradigm, which is currently associated more with companionship, or even agents, where there\u2019s an emphasis on autonomy,\u201d said Jang. \u201cI\u2019ve been thinking of [AI systems] as instruments for thinking, making, playing, doing, learning, and connecting.\u201d<\/p>\n \ud83e\uddea i\u2019m starting oai labs: a research-driven group focused on inventing and prototyping new interfaces for how people collaborate with ai.<\/p>\n i\u2019m excited to explore patterns that move us beyond chat or even agents \u2014 toward new paradigms and instruments for thinking, making,\u2026<\/p>\n \u2014 Joanne Jang (@joannejang) September 5, 2025<\/a><\/p>\n When asked whether OAI Labs will collaborate on these novel interfaces with former Apple design chief Jony Ive<\/a> \u2014 who\u2019s now working<\/a> with OpenAI on a family of AI hardware devices \u2014 Jang said she\u2019s open to lots of ideas. However, she said she\u2019ll likely start with research areas she\u2019s more familiar with.<\/p>\n This story was updated to include a link to Jang\u2019s post announcing her new position, which was released after this story published. We also clarify the models that OpenAI\u2019s Model Behavior team worked on.<\/p>\n
\n\t\t\t\t\t\t\t\t\t\t\t\t\t|
\n\t\t\t\t\t\t\t\t\t\t\t\t\tOctober 27-29, 2025\n\t\t\t\t\t\t\t<\/p>\n