Thinking Machines Lab: Fixing AI's Randomness Problem

Mira Murati's Thinking Machines Lab is tackling AI's biggest headache: unpredictable answers. With $2 billion in funding, they're aiming to eliminate non-determinism in AI, promising consistent, reliable results for businesses and researchers. Their solution focuses on controlling GPU kernels during AI inference, potentially revolutionizing reinforcement learning and paving the way for custom AI models. Murati, formerly of OpenAI, teases an upcoming product for researchers and startups, hinting at a new era of transparency in AI development. Will they succeed where others haven't? The AI world is watching. Hosted on Acast. See acast.com/privacy for more information.