Did anyone who has any connection with tech not expect this? If you rate a specific conversation, did you not expect this conversation to go into training or at least review?
You are explicitly telling them "this is good / bad"...
My non tech people didn't even know that everyone can edit Wikipedia.
And to be fair to them, my motivation to send quite a lot of personal data to an LLM is shockingly high.
I send medical data to openai while I never ever send any AI provider an Image of me or my wife despite me wanting to do Star Wars montages and Studio Ghibli avatars.
I disabled Gemini Activity history because they told me that they will let humans look over it, but I gave claude access to my laptop.
There is a reason why we have consumer protection. Life doesn't need to be hard by default :D
> did you not expect this conversation to go into training or at least review?
Use it for training/review (RLHF, whatever) and "we load the whole thread, not the specific response or surrounding few prompts, and store it for 5 years" seem completely different things. It's not obvious why being tech savvy would automatically make one aware that the 2nd version is happening.
In Claude Code, even just dismissing the feedback may be considered "feedback". I wrote about it here: https://keydiscussions.com/2025/09/29/dont-even-dismiss-the-... The most common feedback on these posts is always "what did you expect?" B@$%! the reason I signed up for one service over the other is because of promises these companies make on the front end about what they do or don't do with my data.
Is this only about when you explicitly click on a thumbs up/down button and provide a message, or is it also for the nagware "How is claude doing today [1] bad [2] fine [3] good" that shows up after 15 minutes of work in the CLI and and requires literally pressing 1 key while in the prompt input field to "send feedback"?
I always press 3 to just get it out of the way (although I treat anything I send OAI/Anthropic as training data, so I don't really care) but if this little popup counts, I wouldn't be surprised if the vast majority of CC sessions out there are now "feedback training data"
The link in the post is for the API and commercial offerings. With that said, as others have already pointed out, how else could they actually make use of the feedback without the data that the feedback was given on.
> When you provide us feedback via our thumbs up / down button, we will store the entire related conversation, including any content, custom styles, conversation preferences, or model settings, in our secured back-end for up to 5 years. [...] We do not combine your feedback with your other conversations with Claude.
I would expect that, if you turned off the feature where they can use your data, that they would remove those features, so you don't either accidentally click them or fail to understand that it isn't some kind of local-to-your-account mechanism.
Generally (and with Anthropic), enterprise plans aren't used for training / tracking / targetting.
There's a whole spectrum of things they could do from most to least aligned with that as a principle.
For example, they could have a thumbs up/down
- just track the metadata of the conversation (e.g. model used, how far through its context it was, how old the session was)
- track metadata of the user as a whole (how often they use Claude, how many other "active" conversations they have on the go, how much they've tested other models, if this is their first time using new tweaks to params etc)
- have contextual feedback of the chat available for humans to review (alongside the comments you put in the feedback), but not explicitly train on it.
Whereas the option they've chosen is:
> If you explicitly report feedback or bugs to us (e.g. via our thumbs up/down feedback button), or otherwise choose to allow us to use your data, then we may use your chats and coding sessions to train our models.
So literally using your session as training input. They also could've allowed this to be an opt-in on bug reports.
Did anyone who has any connection with tech not expect this? If you rate a specific conversation, did you not expect this conversation to go into training or at least review?
You are explicitly telling them "this is good / bad"...
My non tech people didn't even know that everyone can edit Wikipedia.
And to be fair to them, my motivation to send quite a lot of personal data to an LLM is shockingly high.
I send medical data to openai while I never ever send any AI provider an Image of me or my wife despite me wanting to do Star Wars montages and Studio Ghibli avatars.
I disabled Gemini Activity history because they told me that they will let humans look over it, but I gave claude access to my laptop.
There is a reason why we have consumer protection. Life doesn't need to be hard by default :D
> did you not expect this conversation to go into training or at least review?
Use it for training/review (RLHF, whatever) and "we load the whole thread, not the specific response or surrounding few prompts, and store it for 5 years" seem completely different things. It's not obvious why being tech savvy would automatically make one aware that the 2nd version is happening.
In Claude Code, even just dismissing the feedback may be considered "feedback". I wrote about it here: https://keydiscussions.com/2025/09/29/dont-even-dismiss-the-... The most common feedback on these posts is always "what did you expect?" B@$%! the reason I signed up for one service over the other is because of promises these companies make on the front end about what they do or don't do with my data.
That's very deceiving and should warrant a class action law suit in my opinion
If they don't use the available data, how do they act on the feedback?
Hallucinating what might have been.
As noted in the article, the feedback form says:
> Submitting this report will send the entire current conversation to Anthropic for future improvements to our models. Learn More
So I'm not sure why this is being posted to HN. This is nothing new, and nothing not known by anyone who has ever provided feedback.
Is this only about when you explicitly click on a thumbs up/down button and provide a message, or is it also for the nagware "How is claude doing today [1] bad [2] fine [3] good" that shows up after 15 minutes of work in the CLI and and requires literally pressing 1 key while in the prompt input field to "send feedback"?
I always press 3 to just get it out of the way (although I treat anything I send OAI/Anthropic as training data, so I don't really care) but if this little popup counts, I wouldn't be surprised if the vast majority of CC sessions out there are now "feedback training data"
Is this also the case when you answer that how is Claude doing question in Claude Code?
Yes according to the following:
https://privacy.claude.com/en/articles/10023580-is-my-data-u...
The link in the post is for the API and commercial offerings. With that said, as others have already pointed out, how else could they actually make use of the feedback without the data that the feedback was given on.
That's why I never give feedback, ensuring Anthropic has no access to my Claude chats.
I always press 0 (dismiss), but is there a Way to stop prompting me to feedback ?
And how do you know that, just dismissing it isn't considered "feedback".
you can disable telemetry, but that also disables remote control which sucks
I complain at Claude all the time, but do not submit feedback becuase of this exact reason, and because I knew it I assumed everyone knew it...
Giving chat feedback gives Anthropic even more of your data
User: Hey Anthropic, this response was terrible.
Anthropic: Really? Let me read it and see why.
User: How dare you look at my data!!!
... what else would we expect it to do?
> When you provide us feedback via our thumbs up / down button, we will store the entire related conversation, including any content, custom styles, conversation preferences, or model settings, in our secured back-end for up to 5 years. [...] We do not combine your feedback with your other conversations with Claude.
I would expect that, if you turned off the feature where they can use your data, that they would remove those features, so you don't either accidentally click them or fail to understand that it isn't some kind of local-to-your-account mechanism.
I find the lack of privacy forward thinking from a company I pay a couple hundred a month to kind of atrocious. They're buying no loyalty.
Generally (and with Anthropic), enterprise plans aren't used for training / tracking / targetting.
There's a whole spectrum of things they could do from most to least aligned with that as a principle.
For example, they could have a thumbs up/down
- just track the metadata of the conversation (e.g. model used, how far through its context it was, how old the session was) - track metadata of the user as a whole (how often they use Claude, how many other "active" conversations they have on the go, how much they've tested other models, if this is their first time using new tweaks to params etc) - have contextual feedback of the chat available for humans to review (alongside the comments you put in the feedback), but not explicitly train on it.
Whereas the option they've chosen is:
> If you explicitly report feedback or bugs to us (e.g. via our thumbs up/down feedback button), or otherwise choose to allow us to use your data, then we may use your chats and coding sessions to train our models.
So literally using your session as training input. They also could've allowed this to be an opt-in on bug reports.
Well, again, how did people expect that to work
> - just track the metadata of the conversation
This is useless
"So here's how ThanosCorp isn't living up to the altruistic spirit of the ToS for their Human Extinction Project..."
and sky blue
Sophisticated technology experts shocked to find world's 4th largest data-ingestion machine ingesting their data.
It is pretty funny when you think about it. The shock and awe of it all...