OpenAI Unleashes AI-Powered Browser to Challenge Chrome

OpenAI has unveiled its AI-powered web browser, ChatGPT Atlas, opening a new front in the search battle between the company and its rival, Google.

On ChatGPT Atlas, the first tab is the chatbot, but users can also type in a URL or the name of a website, and ChatGPT will direct them to it. When on a site, users can click the “Ask ChatGPT” sidebar to let the chatbot interact with the page’s content. From there, ChatGPT can edit documents, book flights, and summarize film reviews.

Integrating ChatGPT directly into the browser gives OpenAI more control over the experience, allowing the chatbot to scan documents, read pages, and provide assistance without the need for copy-paste or screenshots. With Agent Mode, users can assign the chatbot specific tasks — such as managing a calendar or email — although this feature will initially be limited to Plus and Pro users.

“We think AI represents a rare, once-a-decade opportunity to rethink what a browser can be and how we can most productively and pleasantly use the web,” said OpenAI CEO Sam Altman during a livestream of the announcement. “Tabs were great, but we haven’t seen much browser innovation since then, so we got very excited about the opportunity to rethink what this could be.”

Atlas is available for download on macOS first, with Windows, Google Play, and Apple App Store versions coming soon. 

Finally some competition for web browsers 

The dominance of Google Chrome on both Windows and macOS likely influenced this decision, as OpenAI and others would have faced an uphill battle integrating their chatbots as deeply into Chrome as Google’s Gemini. According to StatCounter, Chrome holds a 71% browser market share, followed by Safari at 13.9% and Microsoft Edge at 4.6%.

At the same time, the success of ChatGPT may have spurred OpenAI to deliver another blow to Google, which has used its browser as a walled garden to promote other services such as Search, Gmail, Drive, and Maps. Google recently added Gemini to Chrome, a move seen as risky given the company’s recent escape from harsher penalties in its 2025 US antitrust case, one of which could have required it to spin off or sell Chrome.

OpenAI is not the only generative AI company developing its own browser. Earlier this month, Perplexity launched Comet, touting similar advantages such as assistance with tasks and information summarization. Mozilla also recently added Perplexity as a search option in Firefox, though its functionality is more limited. Web browsers Opera and The Browser Company have also announced plans to integrate AI.

Other major players, including Anthropic, Meta AI, and Grok, are unlikely to be far behind. Anthropic launched a pilot Chrome extension in August that lets its chatbot, Claude, view web pages and perform tasks autonomously.

Whether any of these efforts make a meaningful dent in Chrome’s dominance remains to be seen. Microsoft saw a temporary uptick in downloads when it integrated ChatGPT into Edge, but usage has since waned, with many users returning to Chrome.

From a browser to math: In a very public lesson on the difference between “solving” a problem and simply “finding” the answer, OpenAI’s latest model, GPT-5, has stirred up controversy.