Malvertising Campaign Abuses Google Ads and Claude.ai Shared Chats to Distribute macOS Infostealer

Date: 2026-05-27
Tags: malware, phishing

Executive Summary

An active malvertising campaign is abusing Google sponsored search results and Claude.ai's shared chat feature to deliver macOS infostealer malware to users searching for Claude downloads on Mac. Both attacks point to the real claude.ai domain because the malicious content is hosted inside Claude's own shared chat feature, and standard advice to verify the destination URL before clicking provides no protection here.

Campaign Summary

Field	Detail
Campaign / Malware	Claude.ai Shared Chat Malvertising Campaign
Attribution	Unknown (confidence: low)
Target	macOS users searching for Claude installation; organizations with employees using Claude on Mac
Vector	Malvertising via Google Ads pointing to malicious Claude.ai shared chat; geographic targeting of Western users
Status	active
First Observed	2026-05-11

Detailed Findings

macOS users searching for Claude downloads through Google are the primary targets, and the geographic filtering in one variant skips CIS-region machines, suggesting selective targeting of Western users. Organizations where employees use Claude on Mac devices face indirect risk if corporate credentials or session tokens are stored in browser profiles or macOS Keychain on affected machines. The use of shared AI platform chats as malware delivery infrastructure has now been documented across Claude, ChatGPT, and Grok, establishing this as a recurring and maturing attack pattern.

MITRE ATT&CK Mapping

Technique	ID	Context
Phishing: Spearphishing Link	T1566.002	Malvertising redirecting to malicious shared chat
Masquerading	T1036.005	Impersonating official Claude Code CLI via shared chat
Ingress Tool Transfer	T1105	Downloading and executing malware payload on macOS

IOCs

Domains

claude.ai

Full URL Paths

_No specific IOCs published; attack uses legitimate claude.ai domain for shared chat delivery_

Splunk Format

"claude.ai"

Detection Recommendations

Monitor Claude.ai shared chat platform for suspicious installation instruction content. Implement email and DNS filtering for suspicious search result redirects from Google Ads. Educate users that legitimate AI tool downloads come only from official vendor documentation, never from search-result installation instructions. Implement endpoint detection for macOS processes spawning from browser contexts with unusual privilege escalation patterns. Monitor for lateral movement from compromised developer endpoints that may have access to corporate credentials.

References

[Security Boulevard / Evan Rowe (CISO Whisperer)] Attackers Abuse Google Ads and Claude.ai Shared Chats to Push Mac Malware (2026-05-11) — https://securityboulevard.com/2026/05/attackers-abuse-google-ads-and-claude-ai-shared-chats-to-push-mac-malware/