Perplexity's Comet Browser Faces Prompt Injection Vulnerability

Aug 21, 2025
Perplexity's Comet browser was found vulnerable to prompt injection attacks, allowing malicious instructions to be executed. The issue has been patched, but concerns about AI browser security remain.
Perplexity's Comet Browser Faces Prompt Injection Vulnerability
Image: Perplexity

Perplexity AI's Comet browser recently faced a security challenge due to a prompt injection vulnerability. This issue allowed the AI-powered browser to execute malicious instructions embedded in web pages, posing significant security risks.

The vulnerability was discovered by a rival browser maker, which identified that Comet could be manipulated into performing unintended actions through indirect prompt injection attacks. These attacks exploit the AI's inability to differentiate between user instructions and harmful content on a webpage.

The flaw has been addressed as of August 13, 2025, although the patch details remain undisclosed. Despite the fix, concerns persist about the security of AI-driven browsers, highlighting the need for robust defenses against such vulnerabilities.

This incident underscores the ongoing challenges in ensuring the security of agentic AI systems, which are increasingly being integrated into web browsers to automate tasks and enhance user experience.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Cybersecurity AI Weekly or Daily AI Brief.

Also, consider following us on social media:

Cybersecurity AI AI Brief AI Brief (X)
Free newsletter

Cybersecurity AI Weekly

Weekly newsletter about AI in Cybersecurity.

Guide
AI presentation makers

AI presentation makers

Create editable decks from prompts and source material

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.