 "How OpenAI’s ChatGPT Atlas browser will be different from Chrome and others")
OpenAI has entered a new phase in its evolution with the introduction of ChatGPT Atlas, a standalone web browser built entirely around the company’s flagship artificial intelligence chatbot.
The launch, announced on Tuesday, signals OpenAI’s intention to transform the way users explore the internet — and to challenge the market dominance of Google Chrome, which has long set the standard for web browsing.
The unveiling of Atlas follows months of speculation about OpenAI’s next big step after the success of ChatGPT, which now attracts over 800 million weekly active users.
With Atlas, OpenAI seeks to integrate conversational AI into everyday web navigation, research, shopping, and productivity, marking a shift from the traditional search model dominated by keyword-based results.
During a livestream announcement, OpenAI CEO Sam Altman described the moment as an opportunity to rethink what web browsers could be in an AI-driven era.
“We think that AI represents a rare, once a decade opportunity to rethink what a browser can be about,” Altman said.
“Tabs were great, but we haven’t seen a lot of browser innovation since then.”
The rollout of ChatGPT Atlas begins with macOS users worldwide, with Windows, iOS, and Android versions expected soon.
The browser will be free to use, though some of its most advanced features — including the new “Agent Mode” — are limited to ChatGPT Plus, Pro, and Business subscribers.
The announcement immediately drew attention across the tech industry and the markets. Alphabet, Google’s parent company and owner of Chrome, saw its shares dip by 1.8 per cent following the news.
How ChatGPT Atlas is different from other browsers
The arrival of ChatGPT Atlas takes place amid what many are calling the “AI browser wars.” Over the past year, a number of companies have launched or upgraded browsers with AI capabilities in a race to redefine how people search and interact online.
Competitors like Perplexity’s Comet, Arc, Opera’s Neon, and Brave Browser have already added conversational features that can summarise web pages, assist with writing, or automate basic tasks.
Tech giants such as Google and Microsoft have been embedding AI into their browsers — Gemini AI in Chrome and Copilot in Edge — to maintain relevance as user behaviour shifts toward conversational search.
In this context, OpenAI’s Atlas is notable not because it adds AI to a browser, but because it builds the entire browsing experience around ChatGPT. This approach contrasts with Chrome and Edge, which still centre on conventional search results enhanced by AI elements.
OpenAI’s Engineering Lead, Ben Goodger, pointed out that ChatGPT is fundamental to Atlas’s design philosophy.
“ChatGPT is core to the company’s first browser,” he said during the livestream. By embedding ChatGPT directly into every aspect of web navigation, OpenAI hopes to make browsing more interactive, personal, and productive.
Despite the growing number of AI-focused browsers, Chrome continues to dominate with 71.9 per cent of the global market share as of September, according to StatCounter.
Yet, analysts say OpenAI’s entry could introduce meaningful competition — particularly among users who already rely heavily on ChatGPT for work, study, or research.
How ChatGPT Atlas uses AI at the core
ChatGPT Atlas does not simply bolt an assistant onto an existing browser. Instead, it reimagines the browser as an AI workspace, where every tab, field, and page can become part of a conversation.
When users open a new tab, they can immediately ask ChatGPT a question, paste a link, or begin a complex task such as planning a trip or writing a report.
The browser’s integrated interface lets users switch between dedicated tabs for search, images, videos, and news, all guided by ChatGPT’s assistance.
OpenAI’s Lead Designer Ryan O’Rouke explained how Atlas changes the search process. If a user types a query such as movie reviews, Atlas presents a chatbot-style summary first, followed by traditional search results including links, images, and videos.
This reverses the structure of Google search, where AI insights appear alongside results rather than defining them.
“We’ve made some major upgrades to search on ChatGPT when accessed via Atlas,” O’Rouke said during the livestream. The system aims to deliver immediate, conversational answers while still giving access to standard web pages for deeper exploration.
This design effectively turns the AI response into the primary browsing interface, positioning traditional web results as supplementary. The layout and experience are intended to feel like an ongoing dialogue between the user and ChatGPT, rather than a static list of pages.
Atlas also introduces a ChatGPT sidebar that can be opened on any website. From there, users can request article summaries, compare prices, analyse data, or receive contextual explanations of what they are viewing.
This means the AI is not only aware of what page the user is on, but can engage with it in real time — providing an experience more dynamic than traditional browser add-ons or extensions.
What we know about ChatGPT’s Agent Mode
Among Atlas’s most advanced features is Agent Mode, which transforms ChatGPT from a conversational assistant into an autonomous task performer. In this mode, the AI can browse websites, fill out forms, make comparisons, or even complete purchases without manual clicks.
During the product demonstration, OpenAI engineers showcased how ChatGPT could locate a recipe online, navigate to Instacart, and automatically add all required ingredients to the shopping cart — completing the transaction-ready workflow independently.
Agent Mode is initially available only to ChatGPT Plus, Pro, and Business subscribers. It builds upon OpenAI’s earlier experiments with agent tools inside ChatGPT, but integrates those capabilities directly into the browser for smoother operation.
The technology aims to reduce repetitive digital tasks — like booking flights, comparing product specifications, or researching destinations — into a single natural-language command.
While early testing by technology reviewers has shown that AI agents can sometimes be slow or make small errors, OpenAI is positioning this feature as the foundation of future browsing automation.
How ChatGPT Atlas will ‘remember’ your work
Another cornerstone of ChatGPT Atlas is its browser memory system, which allows the AI to recall pages, notes, or projects from past sessions.
This memory works similarly to ChatGPT’s existing long-term memory feature but is tailored for browsing activity.
The browser can remember which websites a user visited, what topics they researched, and what tasks were started but not completed.
This allows Atlas to offer more relevant suggestions — such as reopening a travel itinerary or referencing a previously read article — without the user having to search again.
Adam Fry, OpenAI’s Product Lead, said that this feature adds a new level of continuity and personalisation to browsing. Users can delete, edit, or view all stored memories and can opt out entirely if they prefer not to have any browsing data remembered.
OpenAI has stated clearly that data collected through Atlas is not used to train AI models, and that privacy controls remain in the user’s hands. The browser includes full incognito mode, parental settings, and visibility tools to show what is being stored.
When memory is disabled, Atlas behaves like a standard browser, but when turned on, it behaves more like an assistant that remembers context over time.
The idea is to allow users to resume tasks seamlessly — whether that means continuing research, drafting a document, or planning a trip.
The memory system also supports voice and natural language commands, letting users manage their browsing experience conversationally. For instance, you can say, “Reopen the travel site from yesterday” or “Close my recipe tabs,” and ChatGPT will execute those commands directly.
How ChatGPT Atlas will affect your productivity
ChatGPT Atlas also focuses heavily on improving writing and task management.
The browser incorporates in-line editing and composition tools, allowing users to write or refine content directly in web fields — such as replying to emails, filling job applications, or editing shared documents — without leaving the page.
This approach eliminates the need for copying and pasting text between ChatGPT and other platforms, one of the most common user frustrations.
With Atlas, ChatGPT’s capabilities are embedded throughout the web, meaning its grammar correction, summarisation, and rewriting tools are always accessible in context.
The sidecar feature, which has become a hallmark of AI browsers, is also included in Atlas. This lets ChatGPT automatically interpret the content on your screen — including text, links, and forms — without requiring manual input.
The feature removes the need to provide context manually, saving time and streamlining user workflows.
OpenAI claims this design will make online interactions more efficient, whether for professionals managing complex workflows or casual users browsing news and social media.
How Google, Microsoft, and the browser industry reacted
OpenAI’s browser arrives at a time when Google and Microsoft are heavily investing in AI integration to protect their own ecosystems.
Google has begun embedding AI-generated overviews within its search results, offering summaries alongside traditional links. It also introduced the Gemini AI model into Chrome for US users, with plans to expand to mobile platforms.
Chrome now includes a “sparkle” icon that opens its built-in chatbot interface — a direct response to user demand for conversational interaction within search.
Microsoft, on the other hand, has continued to enhance its Edge browser with Copilot, which acts as a companion assistant capable of summarising pages, generating text, and organising research.
Despite these upgrades, Chrome remains the market leader, commanding a 71.9% share globally. Microsoft Edge holds a modest but stable second place, while Safari, Firefox, and others fill smaller niches.
In September, US District Judge Amit Mehta ruled that Google was not required to divest its Chrome browser and could continue paying partners to promote its search engine.
The judge noted that the rapid evolution of generative AI across the industry posed enough competitive pressure to justify those agreements.
Still, analysts suggest that OpenAI’s move into browsing could redirect user engagement and advertising revenues if Atlas gains traction.
By making ChatGPT the starting point for information retrieval rather than Google Search, OpenAI could shift the flow of web traffic and redefine how online advertising and content discovery work.
With inputs from agencies
 "Why Trump has cancelled his talks with Putin in Budapest")
 "Belgian court closes Mehul Choksi’s last escape route, orders extradition to India for PNB scam")
 "Meet the brave 6 Operation Sindoor heroes awarded the Vir Chakra")
 "Magnus Carlsen criticises Vladimir Kramnik after Naroditsky's death: 'He was not in a great space'")
 "Why Trump has cancelled his talks with Putin in Budapest")
 "Belgian court closes Mehul Choksi’s last escape route, orders extradition to India for PNB scam")
 "Meet the brave 6 Operation Sindoor heroes awarded the Vir Chakra")
 "Magnus Carlsen criticises Vladimir Kramnik after Naroditsky's death: 'He was not in a great space'")
