Mira Murati, former Chief Technology Officer at OpenAI, has unveiled her new venture, Thinking Machines Lab. The startup aims to develop artificial intelligence systems that prioritize human-AI collaboration rather than focusing solely on autonomous AI development.
The company launches with approximately 30 employees, including several former OpenAI colleagues. Key leadership positions are filled by Barret Zoph as CTO and John Schulman as chief scientist, both previously associated with OpenAI.
Core Focus Areas
According to company statements, Thinking Machines Lab will concentrate on three main objectives:
- Helping users adapt AI systems to their specific needs
- Developing foundations for more capable AI systems
- Promoting open science to improve industry-wide understanding
The startup emphasizes multimodal capabilities, which allow for more natural communication between humans and AI systems. Unlike other AI companies founded by former OpenAI executives, which focus on developing superintelligent AI, Thinking Machines Lab aims to bridge the gap between AI capabilities and human usability.
While specific products and funding details remain undisclosed, the company has expressed its commitment to transparency in its work, though this doesn’t necessarily mean open-sourcing their models. The startup acknowledges current limitations in AI interaction, noting that while chatbots excel at answering questions, they often struggle with refinement and customization.
The company plans to maintain high safety standards and engage in collaborative efforts with the broader AI community through technical publications and code sharing.