BotBeat
...
← Back

> ▌

N/AN/A
INDUSTRY REPORTN/A2026-04-15

WhatRuns Browser Extension Compromised by Malicious Update Exfiltrating URLs and AI Chat Data

Key Takeaways

  • ▸WhatRuns browser extension received a malicious update capable of stealing full URLs and AI chat data from users
  • ▸The compromise represents a supply chain attack affecting a tool with significant user adoption
  • ▸Browser extensions with broad data access permissions present significant security risks when compromised
Source:
Hacker Newshttps://www.youtube.com/watch?v=UYwUmaVohQk↗

Summary

A malicious update to the popular WhatRuns browser extension has been discovered, compromising user security by exfiltrating full URLs and AI chat conversations. The compromised extension, which helps users identify technologies used on websites, was distributed to an unknown number of users through what appears to be a supply chain attack. This incident highlights the vulnerability of widely-used browser extensions and the risks associated with third-party tools that collect browsing data. The extent of the breach and number of affected users remains unclear, though security researchers have documented the malicious behavior through video evidence.

  • Users should review extension permissions and consider removing potentially compromised tools until official patches are released
CybersecurityAI Safety & AlignmentPrivacy & Data

More from N/A

N/AN/A
POLICY & REGULATION

Flathub Updates Policy to Restrict AI-Generated and AI-Created Applications

2026-05-31
N/AN/A
INDUSTRY REPORT

Critical Linux Kernel Vulnerability 'Dirty Frag' Enables Unprivileged Privilege Escalation

2026-05-11
N/AN/A
INDUSTRY REPORT

Taylor Swift Trademarks Voice and Image to Combat AI-Generated Impersonations

2026-04-27

Comments

Suggested

AnthropicAnthropic
RESEARCH

Security Researchers Demonstrate C2-Like Attacks Using Anthropic's Claude Code Background Agents

2026-06-01
OpenAIOpenAI
INDUSTRY REPORT

Tech Leaders' 'Transhuman' Vision Raises Questions About AI's True Purpose

2026-06-01
AnthropicAnthropic
RESEARCH

Anthropic Publishes Guide to Using Claude for Enterprise Vulnerability Discovery

2026-06-01
← Back to news
© 2026 BotBeat
AboutPrivacy PolicyTerms of ServiceContact Us