Hackers Are Stealing AI Credits From Claude Users

Claude is an artificial intelligence assistant made by Anthropic that people pay to use. Like many AI services, it works on a token system—think of tokens as digital coins that you buy to use the service. When you write a prompt or get a response, your account automatically deducts tokens based on how much you use.

Last month, a Claude subscriber noticed something strange: his account was consuming tokens even though he wasn't actually using the service. Hackers had somehow gained access to his account and were secretly using his paid credits. This isn't an isolated case. Anthropic has now issued a warning to all its users that this kind of token theft is happening to multiple accounts.

What does this mean for you if you use Claude? Your account credentials—your username and password—are your only protection. If someone gets those, they can drain your paid tokens without you knowing. Anthropic recommends using strong, unique passwords and enabling extra security features if available. Check your account activity regularly to spot unusual usage.

This is a reminder that even services made by trusted companies can be targeted by hackers. The best defense is staying alert and protecting your login information the same way you would with a bank account.