OpenAI Shuts Down Atlas, But AI Browser Ambitions Expand
▼ Summary
– OpenAI is sunsetting its AI-powered browser Atlas and redistributing its agentic features into ChatGPT’s desktop app and a new Chrome extension.
– The shutdown follows OpenAI’s CEO of applications telling the team to cut back on “side quests,” which also led to the closure of its AI video tool Sora.
– OpenAI concluded that the browser is a feature, not a destination, and is folding Atlas’s capabilities into places people already work, including Chrome.
– OpenAI is launching a Chrome extension that gives ChatGPT access to webpage context, allowing users to ask questions, summarize content, or start tasks from the browser.
– The ChatGPT desktop app is getting a more robust browser for browsing websites and interacting with pages, plus a cloud browser for AI agents to complete tasks on a user’s behalf.
OpenAI is quietly pulling the plug on Atlas, its experimental AI-powered browser that launched just last October with ChatGPT deeply embedded inside. But rather than abandoning the vision of AI-enhanced web browsing, the company is redistributing Atlas’s most promising agentic features into the tools users already rely on , namely ChatGPT’s desktop application and a new Google Chrome extension.
The decision to sunset Atlas follows a directive from OpenAI’s applications CEO, Fidji Simo, who earlier told the team to eliminate what she called “side quests.” That internal pivot already led to the shutdown of Sora, OpenAI’s AI video generation tool. Now, Atlas joins the list of casualties in the company’s effort to streamline its focus.
For much of the past year, the broader AI industry waged a competitive battle to unseat Google Chrome as the dominant gateway to the web. Perplexity launched its Comet browser, The Browser Company debuted Dia, and both Google and Microsoft infused Chrome and Edge with new AI-powered capabilities. After months of experimentation, however, OpenAI appears to have concluded that the browser itself is a feature, not the ultimate destination for users. The company is now folding Atlas’s browser-like agent functions into the environments where people already spend their time , and that prominently includes Chrome itself.
OpenAI is rolling out a ChatGPT extension for Chrome that grants the AI access to the context of whatever page a user is viewing. From there, users can ask questions about the webpage, request summaries, or initiate longer, more complex tasks , all without leaving the browser. This move positions ChatGPT as a direct rival to Google’s Gemini Side Panel, which offers a similar suite of in-browser capabilities.
At the same time, OpenAI is upgrading its ChatGPT desktop app with a more powerful integrated browser. This feature enables users to browse websites, log into accounts, download files, and interact with web pages directly within the ChatGPT interface. Beyond that, a separate cloud-based browser running on OpenAI’s own servers allows the app’s agents to complete tasks autonomously on a user’s behalf.
Taken together, these updates transform ChatGPT into a continuous, cross-platform workspace that spans Chrome, the desktop app, and an AI agent. The message is clear: OpenAI is betting that the future of browsing isn’t a new browser at all, but an intelligent layer woven seamlessly into the tools people already use.
(Source: TechCrunch)
