Jarvis is real. Google’s new AI prototype accidentally found its way to the public through the Chrome Web Store. This isn’t just some AI helper that can tell you the weather or remind you about doctor appointments. The Information reported that the store page describes the Jarvis prototype as “a helpful companion that surfs the web for you” through web browsers to take care of common tasks such as buying groceries, booking flight and research topics

In other words, Google’s new AI has the ability to take control of a computer to complete these simple tasks without manual human input. The prototype that popped up on Google’s extension store wasn’t in full working mode. A reporter tried to use Jarvis but access permissions prevented the app from performing any functions.

Google took down Jarvis’ store page by midafternoon ahead of its planned reveal in December. Jarvis will have some competition from outfits like Anthropic and its Claude AI assistant that started its public beta phase last month. Anthropic’s AI assistant can do more than just basic functions through a web browser. Claude can take over a computer to type text, activate buttons and move the cursor.

This article originally appeared on Engadget at https://www.engadget.com/ai/google-accidentally-leaked-a-preview-of-its-jarvis-ai-that-can-take-over-computers-203125686.html?src=rss