Why can't ChatGPT Voice set a timer? | Voice AI expert explains

A voice assistant saying it has started a timer is not evidence that it is measuring elapsed time. In Julia Turc's interview and demonstration, examples from OpenAI's ChatGPT produce inconsistent timing claims. Turc then describes building an assistant with explicit start-and-stop functions, letting software maintain the timer while the model interprets the spoken request.

That segment, sponsored by Mistral AI, illustrates a concrete division of responsibility: language understanding selects an operation, and a tool performs the measurement. The selected examples support that design point; they do not establish how frequently a particular voice product fails or provide a controlled comparison between providers.

“You think it's more likely that I'm a superhuman speed traveler than you got the time wrong.”

Julia Turc

“However, they're LMs at the end of the day, they just don't have this concept of time.”

Juman Mal

“Look, we put a man on the moon, so I'm pretty sure we can build a voice agent that sets a timer.”

Julia Turc

“the fact that you can't really control these LLMs very well, it just exposes a lot of potential dangers”

Juman Mal

“we have loads of engineers who will solve this problem. And that sounds like a solution. It's not an actual solution because they haven't solved the thing”

Juman Mal