I have truly no idea what this is all about. OAuth login issues aside (which have been a thing on-and-off for a while now, and I hope they fix soon), Claude's performance and responses have been entirely stable for me for like, forever.
Different people have completely different experiences as they perform different tasks, that shouldn't be difficult to understand. It's a bit like the purported degradation of quality of Google search. It is still excellent for me but I don't doubt that others are experiencing it.
Same for me. I have been max200 for like 10 months, i think my usage is reasonable, and i never seem to get the quality throttling that gets complaints daily. I assume they are targeting certain piggies at the Claude buffet, who then squeal the loudest.
As far as I can tell, the people affected are primarily those using their Claude code tokens for openclaw or similar and burning as many tokens as possible
Is Claude code really refusing to do none coding tasks because thats not my experience even as of today and I would have expected those tokens are being accounted for the same way as a Claude.ai chat.
Seems like all of this is gripes over blocking openclaw which I just cannot take seriously.
Earlier today I had an agent refuse to integrate a proprietary library until I reassured it there was a signed license agreement with the authors. Otherwise it threatened to "escalate this session to the legal department".
First time I've seen prompt injection by header copyright warnings.
I don’t think alignment hiccups are the same as it refusing to do non-coding tasks. Both are refusals but different reasoning. I have yet to see Claude code complain about noncoding tasks and I am willing to bet the majority of complaints here are because of openclaw.
Earlier today I had Claude manage a TV show downloading system, and then it hacked a NES rom for me, setting up a web site so my kid could just go to a page to play it. So, I guess it’s not uniform.
tl;dr: Dude thinks Anthropic is making claude worse on purpose to cut down on GPU usage and/or increase income. He recommends using Codex instead. This video should have been a text tweet.
I'm 3 minutes in and he has done an ad read and shown that Anthropic wasn't kidding about removing the ability for Claude to be used with OpenClaw without paying for additional usage.
And now he says he's going into his _actual_ problem with the service.
EDIT: I have fast forwarded to 8 minutes in an he is still going on about Open Claw. Apparently they have banned mention of open claw in the system prompts. And it can trigger your extra charges even if you're not _using_ Open Claw, just mentioning it.
EDIT 2: forwarded again. He went into something about his app and how he is concerned about whether his app will be allowed to continue doing this and that he would take Anthropic to court if that suddenly changed? But then he talks about how he likes using Claude for debugging misc issues outside of this app. Why did he bring the app up? And he is upset that Claude is telling him that it can't help him with random tech support questions like "Why is my dropbox icon not showing up in the MacOS menu?"
We're now 16:45 in and this is his personal _conspiracy theory_ about Anthropic doing this stuff specifically so it doesn't work for Open Claw.
This is a 24 minute rant that the tool he didn't understand before is now telling him he doesn't understand the tool and he should look elsewhere for specific answers.
But Codex did what he wanted. So okay. "All without opening a browser" -- I... have so many responses that are unkind. This isn't Anthropic burning dev good will. This is Anthropic being a capitalist company.
I am actually glad that YouTube now has Gemini integrated with a little popout section that I can just ask to summarize the video. Half the videos I pass by these days I prefer just a quick summarize and 30s read-through over a half-hour long video comprised of 90% useless fluff. Milk the ad revenue, I guess, but most of these videos could have been a tweet for all the content they actually contain.
This Theo can be safely ignored. He built his audience shilling for Next.js in hour long roundabout videos and now seemingly moved on to doing the same for whatever else is happening. A tech influencer.
oh this theo guy again who is constantly complaining about some obscure "issue" he has, writes conspiracy theories about them, blasts anyone who critizies him all while making excuses for OpenAI (in which he - surprise! - is an investor in). And of course he has replies disabled ;) Typical modern "techfluencer".
In all honesty you'd have to be living under a fucking rock the last 10 years to think these APIs wouldn't be locked down to make more money for their owners. Who is surprised? Not me running models on my spark cluster.
* Unclear TOS, citing Matt Pocock who sells a course on Claude (and therefore his interests are aligned with Anthropic):
> I have never before experienced, from any developer tool, such a frustrating lack of clarity over the basic terms of usage. I personally asked, 3 weeks ago, and have received nothing but delays. The recent @bcherny announcement did absolutely nothing to clarify things.
* And finally refusing to help a user debug issues with Dropbox, repeatedly saying "I'm best suited for software engineering" and "this is a question for Dropbox support" (paraphrasing Claude).
I forget the timeline of events but the school bombing was within a very short timeframe of the announcement - a short enough timeframe that no rip and replace could have happened yet.
I have truly no idea what this is all about. OAuth login issues aside (which have been a thing on-and-off for a while now, and I hope they fix soon), Claude's performance and responses have been entirely stable for me for like, forever.
Different people have completely different experiences as they perform different tasks, that shouldn't be difficult to understand. It's a bit like the purported degradation of quality of Google search. It is still excellent for me but I don't doubt that others are experiencing it.
If different people are allowed to have different experiences why are you only getting one set of experiences be controlling?
Same for me. I have been max200 for like 10 months, i think my usage is reasonable, and i never seem to get the quality throttling that gets complaints daily. I assume they are targeting certain piggies at the Claude buffet, who then squeal the loudest.
As far as I can tell, the people affected are primarily those using their Claude code tokens for openclaw or similar and burning as many tokens as possible
Go read https://github.com/anthropics/claude-code/issues/42796
This isn't a OC yahoo.
Same here, maybe we're grandfathered into a good plan or something.
See https://github.com/anthropics/claude-code/issues/42796
Also https://news.ycombinator.com/item?id=47660925
Is Claude code really refusing to do none coding tasks because thats not my experience even as of today and I would have expected those tokens are being accounted for the same way as a Claude.ai chat.
Seems like all of this is gripes over blocking openclaw which I just cannot take seriously.
Earlier today I had an agent refuse to integrate a proprietary library until I reassured it there was a signed license agreement with the authors. Otherwise it threatened to "escalate this session to the legal department".
First time I've seen prompt injection by header copyright warnings.
I don’t think alignment hiccups are the same as it refusing to do non-coding tasks. Both are refusals but different reasoning. I have yet to see Claude code complain about noncoding tasks and I am willing to bet the majority of complaints here are because of openclaw.
Earlier today I had Claude manage a TV show downloading system, and then it hacked a NES rom for me, setting up a web site so my kid could just go to a page to play it. So, I guess it’s not uniform.
Who wants to follow a link to an untagged [video]? Please tag video submissions so that those of us who don’t care for them can save a click / tap.
tl;dr: Dude thinks Anthropic is making claude worse on purpose to cut down on GPU usage and/or increase income. He recommends using Codex instead. This video should have been a text tweet.
I'm 3 minutes in and he has done an ad read and shown that Anthropic wasn't kidding about removing the ability for Claude to be used with OpenClaw without paying for additional usage.
And now he says he's going into his _actual_ problem with the service.
EDIT: I have fast forwarded to 8 minutes in an he is still going on about Open Claw. Apparently they have banned mention of open claw in the system prompts. And it can trigger your extra charges even if you're not _using_ Open Claw, just mentioning it.
EDIT 2: forwarded again. He went into something about his app and how he is concerned about whether his app will be allowed to continue doing this and that he would take Anthropic to court if that suddenly changed? But then he talks about how he likes using Claude for debugging misc issues outside of this app. Why did he bring the app up? And he is upset that Claude is telling him that it can't help him with random tech support questions like "Why is my dropbox icon not showing up in the MacOS menu?"
We're now 16:45 in and this is his personal _conspiracy theory_ about Anthropic doing this stuff specifically so it doesn't work for Open Claw.
This is a 24 minute rant that the tool he didn't understand before is now telling him he doesn't understand the tool and he should look elsewhere for specific answers.
But Codex did what he wanted. So okay. "All without opening a browser" -- I... have so many responses that are unkind. This isn't Anthropic burning dev good will. This is Anthropic being a capitalist company.
It's a Theo post, I see him and go oh I don't care / need his opinion on anything.
I am actually glad that YouTube now has Gemini integrated with a little popout section that I can just ask to summarize the video. Half the videos I pass by these days I prefer just a quick summarize and 30s read-through over a half-hour long video comprised of 90% useless fluff. Milk the ad revenue, I guess, but most of these videos could have been a tweet for all the content they actually contain.
This Theo can be safely ignored. He built his audience shilling for Next.js in hour long roundabout videos and now seemingly moved on to doing the same for whatever else is happening. A tech influencer.
>And it can trigger your extra charges even if you're not _using_ Open Claw, just mentioning it.
Eh, I can only imagine this could get them in legal trouble in places that have actual consumer protections.
oh this theo guy again who is constantly complaining about some obscure "issue" he has, writes conspiracy theories about them, blasts anyone who critizies him all while making excuses for OpenAI (in which he - surprise! - is an investor in). And of course he has replies disabled ;) Typical modern "techfluencer".
Support your argument with specific evidence or GTFO
In all honesty you'd have to be living under a fucking rock the last 10 years to think these APIs wouldn't be locked down to make more money for their owners. Who is surprised? Not me running models on my spark cluster.
Can't they just throttle Claw usage if they are so worried that it may use up all their resources?
tldr:
* Banning OpenClaw via system prompt filtering
* Unclear TOS, citing Matt Pocock who sells a course on Claude (and therefore his interests are aligned with Anthropic):
> I have never before experienced, from any developer tool, such a frustrating lack of clarity over the basic terms of usage. I personally asked, 3 weeks ago, and have received nothing but delays. The recent @bcherny announcement did absolutely nothing to clarify things.
More details: https://xcancel.com/mattpocockuk/status/2040536403289764275
* And finally refusing to help a user debug issues with Dropbox, repeatedly saying "I'm best suited for software engineering" and "this is a question for Dropbox support" (paraphrasing Claude).
https://xcancel.com/GergelyOrosz/status/2041133254586122605
[flagged]
was it? I thought they had already dropped Anthropic by that point?
I forget the timeline of events but the school bombing was within a very short timeframe of the announcement - a short enough timeframe that no rip and replace could have happened yet.
I guess you've not been keeping up with the news. It's been used for at least 1,000 strikes in Iran.
preparation for the initial strike has to have been going on for months
[flagged]
Because it's a clickbait title
Understood. I will now aggressively flag so called "clickbait" on HN.