GenTabs is a new experimental product in Google Labs that lets you instantly create web apps with AI. It’s based on the research Google is conducting in Disco, a new downloadable AI product, the company said in a press release Thursday.
Google calls Disco a Gemini-powered «discovery vehicle» that helps reimagine the web, and GenTabs is its first new feature. As with Gemini, or any other AI chatbot, you’ll need to ask it questions to get it to work. Ask GenTabs to help plan a vacation to Osaka, Japan, or learn about the solar system, for example, and GenTabs will generate an interactive web app that helps present that information in a format designed to be visually appealing.
If you’re researching a trip, the web app might include an interactive map showcasing all the top travel spots. Or if you want to learn about outer space, it can include an interactive guide to the solar system, allowing you to zoom in on various planets.
Google, which makes the Chrome web browser, says GenTabs will use your open tabs and chat history to understand the complex tasks underlying the web app it creates.
Don’t miss any of our unbiased tech content and lab-based reviews. Add CNET as a preferred Google source.
Google is starting the feature out with a small group of testers. Its plan is to have Disco features trickle down to other Google products eventually.
The waitlist to download Disco and try GenTabs is open, with signups starting for MacOS users first.
The release of Disco comes as Google begins to pull ahead in the AI race against ChatGPT maker OpenAI. Gemini 3 launched earlier this month to much acclaim, reportedly causing OpenAI CEO Sam Altman to declare a «code red» situation inside the company. Google’s resources and Gemini’s skills put OpenAI in a tough situation as it seeks to become a $1 trillion company.
As the AI race continues, the demand for richer features to attract investor dollars remains high. The problem is that richer features require more computing power, which is expensive to produce, causing electricity prices to go up in certain markets and RAM prices to surge. For a company like Google, which is worth $3.8 trillion, it’s easier to absorb these costs. Also, Gemini 3 was trained on Google’s own Tensor chips, meaning the company doesn’t need to rely on supply from Nvidia, another headache that OpenAI must deal with.