OpenAI, the ChatGPT maker, has unveiled ChatGPT Atlas, an artificial intelligence-powered web browser designed to integrate its popular chatbot directly into the browsing experience.
The company made the browser available on Tuesday, October 21, 2025, for Apple’s MacOS operating system.
The launch marks OpenAI’s move into the web browser market, positioning Atlas as a competitor to Google Chrome, the world’s most widely used browser.
A browser built around AI
According to OpenAI, Atlas is designed to operate without a traditional address bar, centering its functions on ChatGPT to offer users a “super-assistant” experience.
“Today we’re introducing ChatGPT Atlas, a new web browser built with ChatGPT at its core,” OpenAI said in a blog post.
“A browser built with ChatGPT takes us closer to a true super-assistant that understands your world and helps you achieve your goals,” they added
- The browser allows users to interact with ChatGPT directly while navigating websites, enabling context-aware assistance without switching tabs, copying, or pasting information.
- Atlas includes a paid agent mode available for ChatGPT Plus, Pro, and Business users. This feature allows ChatGPT to perform searches, analyze information, and automate tasks based on the user’s browsing context.
“As you use Atlas, ChatGPT can get smarter and more helpful, too. Browser memories let ChatGPT remember context from the sites you visit and bring that context back when you need it,” OpenAI explained. Users can opt in to browser memories, controlling what is stored and deleting history to remove any associated data.
ChatGPT Atlas is currently available worldwide for MacOS users, including Free, Plus, Pro, and Go plans. It is also in beta for Business, Enterprise, and Edu users were enabled by plan administrators. OpenAI plans to expand the browser to Windows, iOS, and Android in the near future.
“Getting started is easy: when you open Atlas for the first time, sign in to ChatGPT and bring your bookmarks, saved passwords, and browsing history with you by importing them from your current browser,” OpenAI said.
Competition with Google
OpenAI’s Atlas comes amid growing competition in AI-powered search and browsing. Google has been adapting to shifting search behaviors since ChatGPT’s debut, introducing AI Mode in search results and integrating its Gemini AI model into Chrome for U.S. users.
Despite OpenAI’s entry, Google Chrome maintains a 71.9% share of the global browser market as of September 2025, according to StatCounter.
A federal judge recently ruled that Google will not have to sell the Chrome browser, allowing the tech giant to continue paying partners to promote its search engine.
“Integrating chat into a browser is a precursor for OpenAI starting to sell ads, which it has yet to do so far. Once OpenAI starts selling ads that could take away a significant part of search advertising share from Google, which has around 90% of that spend category,” said Gil Luria, analyst at D.A. Davidson.
What you should know
The launch of Atlas comes as OpenAI seeks to monetize its AI technology and capitalize on the growing use of large language models for search and productivity. At OpenAI’s DevDay event earlier this month, CEO Sam Altman announced that ChatGPT now has 800 million weekly active users, up from 400 million in February.
- The company has also intensified its growth strategy in 2025, combining record funding and strategic acquisitions to strengthen its position in the AI industry. To enhance its developer offerings, OpenAI announced plans to acquire Windsurf, formerly Codeium, an AI coding platform, in a $3 billion deal to help developers write and debug code using natural language.
- In another strategic acquisition, OpenAI agreed to purchase Statsig, a product-testing startup, in a $1.1 billion all-stock deal, with founder Vijaye Raji joining as Chief Technology Officer of Applications to oversee product engineering and infrastructure.
- To power these ambitious projects, OpenAI also signed a $300 billion, five-year cloud computing deal with Oracle, securing around 4.5 gigawatts of computing capacity to train and run large AI models.
