Would a live product demo be complete without some technical difficulties?
On Tuesday, at its annual DevDay conference in San Francisco, OpenAI revealed its new AI agent called Dots, which, like Meta’s Muse, comes with a cute mascot. Powered by its GPT-6 Astra model, the agents are intended to be “always on,” constantly performing tasks for you in the background.
However, the dots didn’t seem quite lit when OpenAI product team member Holly Li tried to show off her capabilities.
Li calls the AI on her phone like you would a human secretary, and a few rings later, the agent, whom she calls “Dottie,” picks up.
“Can you catch me up just on last night’s user testing?” Li asks.
“I’m checking that out now,” Dots said.
But the AI never responds. Ten agonizing seconds pass with no response, the crowd suspended in awkward silence. Li, while playing, asks the AI how it’s going, and it replies curtly that it’s “always checking.”
“I guess Dottie is having a slow morning,” Li jokes.
When Dot starts its day faster, OpenAI says it will be able to handle tasks in all kinds of professional settings. According to a blog post about the agent, an app developer could use it to monitor user feedback and even create bug fixes themselves. A scientist could ask his personal Dot to learn his research and perform new analyzes as new data comes in – or even automatically insert the new numbers into a paper, in a move we can’t imagine backfiring spectacularly.
Dots represents OpenAI’s latest foray into AI personal assistants and is clearly the company’s answer to Meta’s much-hyped Muse AI agent, right down to its kawaii branding. Muse is imagined as a bean-shaped Labubu, while the Dots are small shapes ready to be fluffed.
But it’s unclear whether OpenAI is aiming for such a broad audience with its product. While Meta has made Muse available for free and marketed it to everyday users on its platforms like Instagram and Facebook, OpenAI is locking Dots behind its $100 per month subscription tier – at least for now. It’s not uncommon for OpenAI to release new features in its paid tiers before making them more widely available, and the brand suggests that’s the route it’s taking.
Learn more about OpenAI: OpenAI cancels next AI model when it shows signs of being evil
Gn bussni
