Thinking Machines Lab, the artificial intelligence startup founded last year by former OpenAI CTO Mira Murati, recently unveiled something called interaction models. At its core, this technology sounds like AI that can interrupt you.

Currently, every AI model you have ever used works in the same way: you speak, it listens; it responds, you listen. Thinking Machines is aiming to change this dynamic by building a model that processes your input and generates a response simultaneously, making the interaction more like a phone call than a text chain. The technical term for this approach is "full duplex."

The company claims its model, named TML-Interaction-Small, can respond in 0.40 seconds. That is roughly the speed of natural human conversation and significantly faster than comparable models from OpenAI and Google. However, this is still a research preview, not a finished product. The company is not releasing it to the public yet. A "limited research preview" is expected in the next few months, with a wider release planned for later this year.

So, what should we make of it? It is hard to say for sure. The benchmarks are impressive, and the underlying concept—that interactivity should be native to a model rather than bolted on afterward—is certainly intriguing. Whether the real-world experience will live up to the technical claims is something we will not know until people can actually use it.