RE: https://neuromatch.social/@jonny/116324676116121930
On reading this thread, I think that Anthropic subscribers who use Claude Code have a very strong case for fraud on the part of Anthropic, given there are multiple redundant - and token-expending - calls to the API baked into this, combined with a lack of ability to choose one's own front-end interface, thus mandating the inefficient and costly expenditures, artificially pumping up usage.
Not to mention there are several things highlighted here which are implemented exactly wrong.
e.g. requiring multiple API calls to generate JSON (instead of having a process that deterministically generates valid JSON - annoying to write, but very possible)
and having a regex for sentiment analysis (instead of calling the LLM that is advertised as capable of performing sentiment analysis)
This entire thing is ass-backwards.
@munin Part of the case on the part of the Claude Code users would have to be demonstrating harm. For example, if the plaintiff could demonstrate that they upgraded from regular ($20/mo) to pro ($200/mo) because of the token over-usage but would have stayed in the lower tier without, then that might be grounds for a case. People that YOLO'd into the top sub and never hit any limits wouldn't really be able to state a claim on which relief could be granted.
Not legal advice, just thinking through how one would have to approach it.
Yes, that is approximately what I was thinking would be suitable grounds.
Also persons who are charged/charging by the hour, given that the extraneous calls would thus constitute "time theft".
Fi 🏳️⚧️
Kevin Neely 