Anthropic warns Claude users sessions reportedly stolen by infostealer malware
Anthropic, the team behind AI chatbot Claude, just let users know about active browser sessions reportedly stolen by infostealer malware.
Hackers used infostealer malware to grab active browser sessions, which let them access accounts and use up users' limits.
Anthropic quickly logged out affected users, revoked those sessions, and even removed saved payment methods to keep things safe.
Infostealer traced to Vidar and RedLine
The attack was traced back to well-known infostealer malware like Vidar and RedLine: these steal browser credentials and cookies from infected devices, making it easy for hackers to skip passwords or two-factor authentication.
The whole thing came up after someone on Reddit said they had downloaded a pirated game before discovering their computer had been compromised.
Anthropic says its own systems were not compromised but urges affected users to scan for malware, reset passwords, and check active sessions. It is also refunding any unauthorized charges while it continues investigating.