Close Menu
Tech Savvyed
  • Home
  • News
  • Artificial Intelligence
  • Gadgets
  • Apps
  • Mobile
  • Gaming
  • Accessories
  • More
    • Web Stories
    • Spotlight
    • Press Release

Subscribe to Updates

Get the latest tech news and updates directly to your inbox.

What's On
I played around with the Pixel 11 Pro Fold. I think Google finally saw the right focus points

I played around with the Pixel 11 Pro Fold. I think Google finally saw the right focus points

16 August 2026
Samsung may be going in for the full iOS-esque Liquid Glass look in OneUI 9.5

Samsung may be going in for the full iOS-esque Liquid Glass look in OneUI 9.5

16 August 2026
Apple is reportedly looking to expand its smart home lineup with smart displays, cameras, and updated HomePods

Apple is reportedly looking to expand its smart home lineup with smart displays, cameras, and updated HomePods

16 August 2026
Facebook X (Twitter) Instagram
Facebook X (Twitter) Instagram
Tech Savvyed
SUBSCRIBE
  • Home
  • News
  • Artificial Intelligence
  • Gadgets
  • Apps
  • Mobile
  • Gaming
  • Accessories
  • More
    • Web Stories
    • Spotlight
    • Press Release
Tech Savvyed
Home»News»Yet another research breaks the hype bubble for AI browsers serving serious security flaws
News

Yet another research breaks the hype bubble for AI browsers serving serious security flaws

News RoomBy News Room3 July 20262 Mins Read
Yet another research breaks the hype bubble for AI browsers serving serious security flaws
Share
Facebook Twitter Reddit Telegram Pinterest Email

AI browsers are being sold as the next big thing. They can summarize pages, book trips, and even make purchases for you. But a new study from the University of Washington found that four of the seven most popular ones come with a security risk serious enough to let malicious websites steal data from other sites you have open. The more capable the browser, the bigger the risk turns out to be.

The 30-year security rule that AI browsers are breaking

Since 1995, every browser has followed a rule called the same-origin policy, which prevents websites from reading each other’s data. If you have your bank open in one tab and visit a sketchy site in another, that sketchy site cannot touch your banking information. AI browsers need to bypass this rule to function, since completing tasks across multiple tabs requires reading across different sites.

That broader access is exactly what attackers can exploit through two methods. The first is prompt injection, where a malicious webpage hides secret instructions that the AI agent follows without realizing it has been manipulated, potentially exposing your private emails, passwords, or calendar details.

The second method is memory poisoning, where planted instructions get stored in the agent’s memory and activate later, even after the original page is closed. Researchers ran a successful proof-of-concept attack on ChatGPT Atlas, demonstrating the risk is real. Claude for Chrome was flagged as particularly risky because its browser extension design lets it inject code directly into webpages.

Which AI browsers are safe and which ones put your data at risk?

Out of seven browsers, ChatGPT Atlas, Chrome with Gemini, Claude for Chrome, and Perplexity Comet were found vulnerable. Microsoft Edge with Copilot, Brave Leo, and Firefox AI Mode showed stronger security properties, though Firefox was also the most limited in capability.

Home page of Perplexity Comet browser on an Android phone.

Researchers disclosed the findings to all companies involved. Anthropic and Firefox did not respond. Whereas Perplexity and OpenAI declined to act, arguing the researchers lacked a complete end-to-end attack demonstration. Meanwhile, Google, Microsoft, and Brave engaged constructively with the findings.

This follows the recent BioShocking exploit, which also showed how AI browsers can be manipulated by context. Right now, the research suggests AI browsers may still be moving faster than their security can keep up.

Share. Facebook Twitter Pinterest LinkedIn Telegram Reddit Email
Previous ArticleClaude Fable 5 is leaving subscriptions, but maybe not for good
Next Article Elon Musk refutes report claiming that an AI device is in development at SpaceX

Related Articles

I played around with the Pixel 11 Pro Fold. I think Google finally saw the right focus points

I played around with the Pixel 11 Pro Fold. I think Google finally saw the right focus points

16 August 2026
Samsung may be going in for the full iOS-esque Liquid Glass look in OneUI 9.5

Samsung may be going in for the full iOS-esque Liquid Glass look in OneUI 9.5

16 August 2026
Apple is reportedly looking to expand its smart home lineup with smart displays, cameras, and updated HomePods

Apple is reportedly looking to expand its smart home lineup with smart displays, cameras, and updated HomePods

16 August 2026
Google may have accidentally leaked a new fitness tracker with a display

Google may have accidentally leaked a new fitness tracker with a display

16 August 2026
AI companies may be gobbling up old books, and I really hope they aren’t destroying them

AI companies may be gobbling up old books, and I really hope they aren’t destroying them

16 August 2026
Claude is getting ambitious with watermarking, and I can smell the problems from a mile away

Claude is getting ambitious with watermarking, and I can smell the problems from a mile away

15 August 2026
Demo
Stay In Touch
  • Facebook
  • Twitter
  • Pinterest
  • Instagram
  • YouTube
  • Vimeo
Don't Miss
Samsung may be going in for the full iOS-esque Liquid Glass look in OneUI 9.5

Samsung may be going in for the full iOS-esque Liquid Glass look in OneUI 9.5

By News Room16 August 2026

Samsung appears to be preparing a major visual overhaul for One UI, with early leaks…

Apple is reportedly looking to expand its smart home lineup with smart displays, cameras, and updated HomePods

Apple is reportedly looking to expand its smart home lineup with smart displays, cameras, and updated HomePods

16 August 2026
Google may have accidentally leaked a new fitness tracker with a display

Google may have accidentally leaked a new fitness tracker with a display

16 August 2026
AI companies may be gobbling up old books, and I really hope they aren’t destroying them

AI companies may be gobbling up old books, and I really hope they aren’t destroying them

16 August 2026
Tech Savvyed
Facebook X (Twitter) Instagram Pinterest
  • Privacy Policy
  • Terms of use
  • Advertise
  • Contact
© 2026 Tech Savvyed. All Rights Reserved.

Type above and press Enter to search. Press Esc to cancel.