Mira Moratti, former CTO of OpenAI, left the AI giant in late 2024 to found her startup, Thinking Machines Lab, raising a staggering $2 billion in less than a year. In her first media interview since founding the startup, Moratti revealed what she was building: “interaction models,” or multimodal AI systems that process voice, text, and video simultaneously and collaborate with humans in near real-time, without waiting for prompts. “I firmly believe that the way to continue building leading-edge AI systems is to do just that [create] “Systems that not only advance autonomously and leave civilization behind, but are more like a tandem bike,” she said on stage at the Bloomberg Technology Summit last week.
“They receive audio, text and video continuously, and deliver output continuously,” Moratti added, describing the model called TML-Interaction-Small, which she plans to release publicly later this year.
Unlike many AI founders who climb traditional computer science or venture capital degrees, Moratti’s path is rooted in product management. She studied mechanical engineering at Dartmouth College, where she worked on building a hybrid race car. After graduating, she joined Tesla in 2013 as a product manager for the Model
In 2018, Moratti joined OpenAI as Head of Applied AI and Partnerships, quickly rising to become Chief Technology Officer. Over the course of six years, I helped guide the deployment of some of the most influential AI products, including ChatGPT, DALL-E, and GPT-4.
During the temporary firing of Sam Altman in November 2023, Moratti served as interim CEO for three days, serving as a bridge between Altman and a small group of executives involved in the leadership change. In hindsight, she said, OpenAI would have collapsed without her intervention. “There was no transition plan, and there wasn’t a lot of thought about transparency, bringing in a team, and providing continuity,” she said. And in the recent lawsuit filed by Elon Musk against OpenAI and Altman, Moratti as well Altman testified He tended to “create chaos.”
Reflecting on her time at OpenAI, she said: “The governance structure in decision-making should not be down to one person. There should be checks and balances.” “To involve as many people as possible, you have to share knowledge, you have to share tools, and that’s part of the reason we took a more open approach with our lab.”
She is now applying those lessons at the Thinking Machines Lab, which has attracted top talent from across the industry. CTO Soumith Chintala joins from Meta. Chief Scientist John Shulman, co-founder of OpenAI, joins from Anthropic. Lillian Wong left her position as Vice President of OpenAI to become a founding member.
At the same time, the company faced talent losses. Meta Superintelligence Labs, led by Scale AI founder Alexander Wang, has done just that Boiled five Founding members. The latest, Joshua Gross, now leads engineering teams at Mark Zuckerberg’s lab. Meta is said to have begun an aggressive recruitment campaign after Moratti Rejected a $1 billion takeover offer In mid-2025.
