Some users are reporting a frustrating pattern: they create a Claude account, enter payment details, the charge goes through, and then they're immediately locked out with a ToS violation notice — all within the same minute.
The sequence reads like a fraud detection system that's too aggressive. The invoice and ban emails arrive simultaneously, suggesting the payment processor and the trust-and-safety system aren't coordinating. You get charged and banned before you've typed a single message.
Anthropics hasn't commented publicly on what triggers these instant bans. Common causes for this kind of automated flagging tend to include VPN usage, device fingerprinting issues, account creation from flagged IP ranges, or geographic mismatches between billing address and IP location. None of those are actual terms-of-service violations — they're fraud signals that also catch legitimate customers.
The problem compounds at the dispute stage. Users banned this way need to contact Anthropic support for reinstatement and their card issuer for a refund if reinstatement fails. Neither process is fast.
Claude competes directly with ChatGPT for paid subscribers, and onboarding friction at the payment step is exactly the kind of thing that pushes people toward alternatives. A system that charges first and asks questions later — with ban enforcement measured in seconds — suggests the fraud thresholds need recalibration on Anthropic's end.