HERMES.md in commit messages causes requests to route to extra usage billing

I find it increasingly ironic that the company that wants you to think software engineering as a profession is doomed, seems to be speedrunning tech fuckups bucket list, most likely using their own product, to achieve this very goal

They're humanists. Haven't you seen those awesome chalk drawings outside their hq?

Do a chargeback?

He should use credit card chargeback.

I am confused.

The person who created the PR is user "sasha-id".

The person saying no to the refund is also user "sasha-id".

What?

Where was it exactly thats someone from Anthropic said no to a refund request? I feel I am missing the obvious somehow.

Claude is running their accounting department

Giving them access to your account or credit card is a bit wild. That's what prepaid cards are for. You charge it with exact amount of money you need to pay for what you want and leave it empty after you pay. You can later watch for bounced payment request to help evaluate their reputation. At this point Anthropic is about as reputable as shady porn site.

there was a time when tech companies gave bug bounties. Now it's fuck you, we vibe coded this slop, and we love it. Oh we emailed your company, ran massive marketing campaigns in the media to pitch replacing you.

I wonder when Anthropic will give refunds for all the sessions with nerfed / dumbed down Opus.

AI company not giving a refund?

I think people put this out of proportion. Yes, you can reason this is ethically correct - I don't object to this. But people used Anthropic, Claude etc... in the first place. Why would you use something to then be disappointed about how it performs, when it comes to AI? Would not be the better and easier strategy to ... not use it in the first place, and make yourself dependable on AI? I don't fully understand this. I would not run into a similar situation because I simply don't use any AI. I actively want to support those folks who don't use AI either - that way we can point out all the ill effects of AI, such as in the case of Anthropic to prioritize on greed.

> Thanks for the report! This was an overactive anti-abuse system. Fixed.

Ah yes, cause who bothers to test any releases to actual paying customers

This is annoying since I have a side project I like to use alchemical names in, and HERMES.md sounds like something I would do. Guess I have to go with AGRIPPA.md, but Hermes Trismegistus is so much cooler...

Searching for the strings of configuration files of other agents in a codebase's git history in order to "detect" unauthorised usage is such a stupid idea I know it 100% came from Claude, and I doubt any of the vibesloppers working at Anthropic bothered to turn their brain on enough for the 5 seconds of thinking it would take to grasp that fact.

They just lost the Claude lottery, that’s all.

I think one day later the guy got his refund? You all need to chill I feel like. HN is a bubble sometimes

https://github.com/anthropics/claude-code/issues/53262#issue...

Another reason to avoid Anthropic products now.

Welcome to the Global Hormuz.

The deeper into the new world order, the more you'll be charged for every breath, by design and by bugs-as-features all the same, refunds be against technofascist manifestos.

Google worked for tens of years to make people disgusted and hating them. Big AI companies succeeded in just a few years, so AI must be an accelerator.

That has a chance to be the highest opportunity cost bug in history ...

"We're already losing literal fuck-tons of money by the minute, so we can't afford to refund you for our mistake."

Another slop coded piece of shit causing stupid bugs.

I can’t believe they paid 100m for some of these employees. They could have bought entire companies of real developers.

[dead]

[dead]

[dead]

C'mon folks, let's stop using Claude|ChatGPT|etc en masse. It's time to start the revolution (from our beds, at least)