Anthropic Users Are PISSED

A lot of people are getting really, really frustrated by Anthropic's latest business practices, and that's charging people more whenever they expect that they're using a harness that Anthropic doesn't want them to use. For example, Hermes or OpenClaw. Here's a tweet from Theo Brown. Fun fact, if you have a recent commit that mentions OpenClaw in a JSON blob, Claude Code will either refuse your request or bill you extra money. This is an empty repo. I'm just calling Claude Code directly. Here's another case as shared from Om Patel here. So some guy was on Claude Max 20x, the $200 a month plan. Yesterday, Claude Code hit him with a "you're out of extra usage" out of nowhere. His dashboard showed 13% weekly usage, 0% current session, but he got $200.98 in extra usage, which should have been covered by his subscription plan. He started binary searching repos and commits until he found the trigger, the string "hermes.md" in a recent git commit message. He then reported it. Anthropic support acknowledged the bug, called it an authentication routing issue, thanked him for finding it, and then refused to refund the money. Here was the actual response: We are unable to issue compensation for degraded service or technical errors that result in incorrect billing routing. Essentially, you can have a subscription plan with Anthropic, pay $100 or $200 a month, use it for coding, and if your code happens to mention OpenClaw or Hermes or one of these harnesses that isn't Claude Code or Claude Cowork, they're gonna either not let you use their product anymore or charge you extra. That's kind of screwed up. There is a slightly happy ending here. Tariq from Anthropic says, sorry, this was a bug with the third-party harness detection and how we pull get status into the system prompt. We're reaching out to affected users and giving them a refund plus another month's worth of credit. So Anthropic did decide to actually refund its users after this sort of became a big deal. But I do wonder if Anthropic would've had this response had this post from An Patel not gotten 1.4 million views and this post from Theo getting 1 million There's some replies here that I kind of have to agree with. It's a good thing Anthropic tells us how ethical they are, or we might worry about their morality, right? They're the ethical AI, yet they have practices where if they don't like that you have the word Hermes or OpenClaw in your code, they're gonna go and then charge you extra for that, or just refuse to like let you use their app.