OpenAI’s new Atlas browser may have some extremely troubling security issues, experts warn – here’s what we know


  • All agentic AI browsers are susceptible to indirect prompt injections
  • Only use agent browsing when you are not handling sensitive information
  • We might have to rethink how browsers work and how we use them

Just days after OpenAI released Atlas, its take on the web browser, the company is struggling to maintain its reputation due to security concerns.

The Chromium-based browser, which has a built-in AI agent for web navigation and automation, has been found vulnerable to indirect prompt injection, which means malicious commands can be hidden in web content to manipulate the agent’s functions.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top