If your Claude subscription sits idle yet your token usage keeps dropping, the cause may not be background tasks — your login session may have been stolen. According to TechCrunch, a wave of Claude paid subscribers are falling victim to "token theft": attackers use infostealer malware to steal login sessions, then drain victims' paid quotas.
Grant De Swardt, an independent AI consultant in East Sussex, U.K., is a textbook case. On August 4, he noticed his Claude Max 20x account ($200/month) burning through tokens on days he wasn't working. The next day he disabled everything connected to Claude — and did nothing — yet usage still climbed. In the clearest controlled test: no work of his own, scheduled Cowork tasks paused or completed, Dispatch/cloud execution off, no active local Claude Code tasks — usage still rose from 45% to 55%. He asked Anthropic for an itemized list; the company didn't provide one, but agreed something was off, suspended his paid account, invalidated all sessions and server-side Claude Code tokens, and refunded £44.49 of the subscription.
Anthropic confirmed to some affected users that malicious actors used common infostealer malware to steal Claude login sessions, then used the stolen session keys to mint unauthorized Claude Code OAuth tokens — either reselling access to the hijacked session or running their own agent workloads on the victim's meter and bill. Such malware (Vidar, LummaC2, StealC, RedLine, Acreed, and Atomic Stealer on macOS) typically harvests already-authenticated session credentials from browsers and local apps — because the session is already authenticated, it walks straight past two-factor verification.
De Swardt's experience is not isolated — similar reports have surfaced on Reddit and GitHub. He described the suspension as wreaking havoc on his business: his job is helping small and mid-size businesses set up agents, so a frozen account meant his core productivity was cut off. Anthropic stresses that the malware did not spread through Claude itself, but from other infections on users' devices. Users who receive such warnings should sign back in, review authorized apps, and revoke unfamiliar Claude tokens. For everyday users, the episode is a reminder: subscription AI quotas are becoming a new target for cybercrime — and the bill can be drained without you noticing.