Gemini’s task automation is here and it’s wild

A few weeks ago, Google and Samsung announced a major development of Gemini on their new devices: task automation. Starting with food delivery and ride-hailing apps, Gemini would be able to use certain apps on your behalf in a virtual window to take care of things like ordering dinner or taking a car to the airport, all based on simple prompts. You know, everything we’ve been promised for years that AI assistants will be able to do. This feature wasn’t available when I started testing the S26 Ultra, but it just arrived in beta as part of an update. And my god, it’s weird to see your phone using itself!

The first prompt I gave was pretty simple: order an Uber to the airport. Gemini asked for clarification on which airport (a good question to ask!) and then did a few steps on its own: adding the destination and choosing to skip the step where you specify your airline, which doesn’t really matter at my local airport since it’s all in one terminal. As promised, the system shut down before the final step and prompted me to review the details before submitting the car request.

A vague and slightly more complicated request to order a coffee and a croissant required a little more input on my part – and a lot of time on Gemini’s part to peruse Starbucks’ hot drink options – but sure enough, he found the flat white on the menu. She was also faced with a crucial decision: order the chocolate croissant warm, or straight from the pastry base? Without my input, he clarified (correctly) that the pastry needed to be reheated. Pretty impressive for an assistant who, just a year ago, was arguing with me over the details of a flight on my calendar.

I still have a lot of testing to do with this automation feature and plan to spend the next few days throwing a few curveballs at it. Still, it’s impressive to see this feature working as expected – at least so far.

Photography by Allison Johnson/The Verge