Yes and when we say things like that we are not talking about plans. Running inference at a profit means api token use is run profitably. It’s a huge unknown what’s happening at the plan level, we know there is subsidy happening but in aggregate impossible to know if it’s profitable or not.
What does this even have to do with the thread? They're hosted there cause it's cheap and extremely easy to do so. Not because it's "specially crafted" for scams.
The harness or the tool is ok but all the defaults as part of the paid pieces of the tool have really bad privacy decisions. So they offer Zen as a pay as you use credit system with access to the models they think work best with the harness. Their own stealth model in it along with a number of the leading new models are always-on sharing data for training purposes. They don’t make this immediately obvious either you have to click through links on their website to see the breakdown.
I am not usually super privacy minded but if you already made it nonobvious this is happening I don’t really trust the underlying tool.
Above is the link. The front page says your privacy is important and says they don’t do training on your data except for the following exceptions which links to this page. Then even their own model is training on your data except there is no opt out. So if you pay for zen and you select one of these models in the tool you have no clue it’s auto training on your data.
Maybe it technically under some regulation runs afoul. The FTC would never bother themselves with this and I don’t believe it’s in the spirit of the intent.
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.
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.
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.
Search and recommendations is terrible. It’s a classic example of what’s wrong at Google.
I used to love watching recommended videos at the end of a video. They were always focused on some extension of the current video watched. Now it’s slop trying to peddle stuff I don’t watch or YouTube shorts. Same with search. It is so terrible, you get maybe 2-4 relevant results and then it again weights all the YouTube slop.
Such a small minority of the customers they want use openclaw and in aggregate a lot of compute use is coming from the total group. Better to stop customers you don’t want. This has zero impact on top line revenue
reply