403
Sorry!!
Error! We're sorry, but the page you were looking for doesn't exist.
The Browser Wars Return: Openai's Atlas Turns The Web Into An Assistant
(MENAFN- The Rio Times) On October 21, OpenAI launched ChatGPT Atlas, a web browser that doesn't just display pages-it reads them and, with permission, acts on them.
A persistent ChatGPT sidebar summarizes articles, compares products, and drafts notes in place. An optional memory feature recalls prior sessions.
Most striking is Agent Mode: the assistant can click through sites, fill forms, and complete tasks like booking travel or checking out on an e-commerce page while showing what it's doing.
Atlas debuts on macOS first, with Windows and mobile promised next; Agent Mode starts on paid plans. The story behind the story is strategic. The battleground has shifted from search boxes to the browser itself-the surface where work actually happens.
To be reliable, AI agents need first-hand access to page context, buttons, and forms, which is hard to deliver as a bolt-on extension. By owning the browser, OpenAI reduces reliance on third-party search and moves directly into Chrome's territory, while meeting Perplexity 's Comet head-on.
Chrome still commands roughly three-quarters of desktop browsing and the largest extension ecosystem, but Atlas and Comet are built for an assistant-first world where“do this for me” replaces“show me links.”
Why this matters, especially for readers outside Brazil who may not track the local tech scene: if agentic browsing works at scale, it compresses time and know-how.
AI Agents Promise Convenience, but Risks Lurk
A café owner in São Paulo filing a municipal form, a freelancer in Nairobi assembling multi-leg itineraries, or a student in Warsaw comparing scholarship rules could hand routine web chores to an assistant and approve the results.
That convenience comes with risks: accuracy lapses, opaque actions, malicious pages that trick agents, and the concentration of power if a few companies mediate what we see and do online.
Expect tighter permission prompts, audit logs of agent steps, and clearer“what happened and why” design to become standard.
What to watch next: Atlas's rollout beyond macOS, whether Agent Mode reaches free users, how publishers and shops respond (from CAPTCHAs to bot policies), and how quickly Chrome weaves comparable automation into a familiar interface.
The winner won't have the flashiest demo, but the agent people trust on slow connections, older laptops, and in languages beyond English.
A persistent ChatGPT sidebar summarizes articles, compares products, and drafts notes in place. An optional memory feature recalls prior sessions.
Most striking is Agent Mode: the assistant can click through sites, fill forms, and complete tasks like booking travel or checking out on an e-commerce page while showing what it's doing.
Atlas debuts on macOS first, with Windows and mobile promised next; Agent Mode starts on paid plans. The story behind the story is strategic. The battleground has shifted from search boxes to the browser itself-the surface where work actually happens.
To be reliable, AI agents need first-hand access to page context, buttons, and forms, which is hard to deliver as a bolt-on extension. By owning the browser, OpenAI reduces reliance on third-party search and moves directly into Chrome's territory, while meeting Perplexity 's Comet head-on.
Chrome still commands roughly three-quarters of desktop browsing and the largest extension ecosystem, but Atlas and Comet are built for an assistant-first world where“do this for me” replaces“show me links.”
Why this matters, especially for readers outside Brazil who may not track the local tech scene: if agentic browsing works at scale, it compresses time and know-how.
AI Agents Promise Convenience, but Risks Lurk
A café owner in São Paulo filing a municipal form, a freelancer in Nairobi assembling multi-leg itineraries, or a student in Warsaw comparing scholarship rules could hand routine web chores to an assistant and approve the results.
That convenience comes with risks: accuracy lapses, opaque actions, malicious pages that trick agents, and the concentration of power if a few companies mediate what we see and do online.
Expect tighter permission prompts, audit logs of agent steps, and clearer“what happened and why” design to become standard.
What to watch next: Atlas's rollout beyond macOS, whether Agent Mode reaches free users, how publishers and shops respond (from CAPTCHAs to bot policies), and how quickly Chrome weaves comparable automation into a familiar interface.
The winner won't have the flashiest demo, but the agent people trust on slow connections, older laptops, and in languages beyond English.
Legal Disclaimer:
MENAFN provides the
information “as is” without warranty of any kind. We do not accept
any responsibility or liability for the accuracy, content, images,
videos, licenses, completeness, legality, or reliability of the information
contained in this article. If you have any complaints or copyright
issues related to this article, kindly contact the provider above.
Most popular stories
Market Research
More Story
Comments
No comment