News Summary

Thinking Machines Lab, a startup founded by former OpenAI CTO Mira Murati, is pioneering an AI model that revolutionizes interactivity by enabling simultaneous processing and response, akin to a phone call rather than a text chain. Their model, TML-Interaction-Small, boasts a response speed of 0.40 seconds, which aligns closely with natural human conversation and outpaces current models from OpenAI and Google. This approach, termed ‘full duplex,’ integrates listening, speaking, and reacting capabilities into a single network, challenging the prevailing turn-based model systems. Although currently in a research preview, Thinking Machines plans a limited release in the coming months, with broader availability later in the year. This innovation could redefine how real-time AI interaction is conceptualized and implemented.