LONDON (IT BOLTWISE) – New AI-powered browsers such as ChatGPT Atlas and Comet promise to revolutionize web browsing. But experts warn of significant security risks associated with the use of these technologies.
Today’s daily deals at Amazon! ˗ˋˏ$ˎˊ˗
The introduction of AI-powered web browsers like OpenAI’s ChatGPT Atlas and Perplexity’s Comet could fundamentally change the way we surf the internet. These browsers offer so-called agentic functions that can perform tasks on behalf of the user by navigating websites and filling out forms. However, these innovative functions also come with significant security risks that many users may underestimate.
Cybersecurity experts warn that AI browser agents pose a greater risk to user privacy than traditional browsers. These agents require extensive access rights to work effectively, including access to users’ emails, calendars, and contact lists. While these features can be useful for simple tasks, they often reach their limits when it comes to more complex tasks and can take a lot of time.
A key security problem with AI browser agents are so-called “prompt injection attacks”. These attacks exploit vulnerabilities by hiding malicious instructions on websites. If an agent analyzes these pages, it can be tricked into carrying out an attacker’s instructions, which could lead to unwanted actions or data leaks.
Companies like OpenAI and Perplexity have already taken steps to mitigate these risks. For example, OpenAI has introduced a “Logged Out Mode” where the agent is not logged into the user account while navigating the web. Perplexity has developed a detection system that can identify prompt injection attacks in real time. Nevertheless, the problem remains a challenge that requires continued attention.
Experts like McAfee’s Steve Grobman point out that the separation between the core instructions of models and the data they process is often unclear, making this problem difficult to solve. Users should therefore exercise caution and ensure that they use strong, unique passwords and multi-factor authentication to protect their accounts.
*Order an Amazon credit card with no annual fee with a credit limit of 2,000 euros! a‿z
Bestseller No. 1 ᵃ⤻ᶻ “KI Gadgets”
Bestseller No. 2 ᵃ⤻ᶻ “KI Gadgets”
Bestseller No. 3 ᵃ⤻ᶻ “KI Gadgets”
Bestseller No. 4 ᵃ⤻ᶻ «KI Gadgets»
Bestseller No. 5 ᵃ⤻ᶻ “KI Gadgets”
Please send any additions and information to the editorial team by email to de-info[at]it-boltwise.de. Since we cannot rule out AI hallucinations, which rarely occur with AI-generated news and content, we ask you to contact us via email and inform us in the event of false statements or misinformation. Please don’t forget to include the article headline in the email: “Security risks with AI-powered browser agents”.
The post Security risks with AI-powered browser agents appeared first on Veritas News.
