On Tuesday, OpenAI unveiled ChatGPT Atlas, its new AI-driven browser, signaling a significant move in its ambition to compete with Google as the primary online information source.
Initially launching on macOS, ChatGPT Atlas will soon expand its accessibility to Windows, iOS, and Android platforms. OpenAI has confirmed that the product will be freely available to all users at the outset.
The browser arena is rapidly emerging as a competitive landscape for AI technologies. While Google Chrome has long been the frontrunner, there’s a growing belief that AI chatbots and digital assistants are transforming how users tackle tasks online. Numerous startups, such as Perplexity’s Comet and The Browser Company’s Dia, are stepping up with their own AI-infused browsers. In response, both Google and Microsoft have begun integrating AI capabilities into their established browsers, Chrome and Edge, to maintain their relevance.
Ben Goodger, OpenAI’s Engineering Lead for Atlas, shared during a livestream on Tuesday that the integration of ChatGPT is central to the functionality of the new browser. Users of ChatGPT Atlas will be able to interact with their search results via a chat interface, akin to features found in Perplexity or Google’s AI Mode.
A standout feature in many AI-centric browsers has been the integrated chatbot located in a side panel, which automatically understands the context of the content displayed. Although this might seem inconsequential, many users routinely find themselves copying text or dragging links into ChatGPT just to provide background information. The sidecar functionality virtually eliminates these challenges, enhancing the overall user experience.
During the livestream, OpenAI’s Product Lead, Adam Fry, confirmed that ChatGPT Atlas will also include this sidecar feature. Additionally, the browser will incorporate “browser history,” allowing ChatGPT to track the websites users visit and their activities, thereby enabling more personalized responses.
AI-enhanced browsers typically feature an AI agent designed to automate online tasks on behalf of users. In early tech tests, we observed that the initial iterations of these web-browsing AI agents often struggled with more complex tasks. While both Perplexity’s Comet and OpenAI’s ChatGPT agent perform adequately for straightforward functions, they can falter when it comes to facilitating bulkier, more complex challenges users may want to delegate to an AI system.
TechCrunch Event
San Francisco
|
October 27-29, 2025
OpenAI’s browser includes a browsing agent as well. Through the “agent mode,” users can instruct ChatGPT to perform minor tasks within the browser on their behalf. However, it’s important to note that agent mode will be initially accessible only to ChatGPT subscribers in the Plus, Pro, and Business tiers.
In an interview at OpenAI’s DevDay conference, ChatGPT’s Head, Nick Turley, discussed his vision for how browsers could redefine the concept of an operating system. He expressed that browsers have transformed the way people accomplish tasks online, and he believes ChatGPT could similarly revolutionize user interactions.
Whether ChatGPT Atlas can significantly challenge Google Chrome’s dominance—which boasts over 3 billion users worldwide—remains uncertain. While AI browsers are generating considerable buzz in Silicon Valley, their widespread impact is yet to be fully realized.
