No current employee of a corporation can promise that data collected today will not in the future be used for any particular purpose. It's simply not up to them, and companies change massively over time.
Get it in an air-tight legal agreement with some kind of audit provision, actual enforcement and penalties, or don't give out data you care about.
You're the human feedback in the RLHF part of Reinforcement learning human feedback. You're also their paypig for their $200 Max plan which you also get limited on. You're also their free advertisement, keep at it. I can't wait for the open-source alternatives to catch up if ever.
This nag often comes up at the very beginning of a session, before you could realistically evaluate whatever A/B test they’re running with any accuracy. I hope they’re not taking the data too seriously.
Despite what they say in that GH comment that it was reduced by 10x my anecdata is that it has been hyper-aggressive in the last few days. I had it popup twice during a single context window session. It makes me feel like they are really concerned about model regressions and in turn their lack of confidence makes me really not confident about what's going on behind the scenes.
I've definitely responded with prompts for feedback with "Stop nagging me for feedback all the time. I just want to (do what the app is for) without interruption."
It's kind of shocking how some people just don't get how insanely insulting it is for an application to constantly ask for feedback.
I had one of these pop up in Android Auto earlier while I was driving. I'd used the Google voice assistant to change the destination I was navigating to, and a few seconds it threw a feedback popup over the top of the map, obscuring the name and distance to the next junction. Very annoying and distracting, potentially unsafe since I was going around 70 mph while surprised by this.
It should at least wait until I've finished my journey and parked up. Not that I'm going to bother giving ad hoc feedback then either.
Hey HN, Boris from the Claude Code team here. Sharing a bit about why we have this survey, how we use it, and how to opt out of it-
1. We use session quality feedback as a signal to make sure Claude Code users are having a good time. It's helpful data for us to more quickly spot & prevent incidents like https://www.anthropic.com/engineering/a-postmortem-of-three-.... There was a bug where we were showing the survey too often, which is now fixed (it was annoying and a misconfiguration on our part).
2. Giving session quality feedback is totally optional. When you provide feedback, we just collect your numerical rating and some metadata (like OS, terminal, etc.). Giving feedback doesn't cause us to log your conversation, code, or anything like that. (docs: https://code.claude.com/docs/en/data-usage#session-quality-s...)
3. We don't train on quality feedback data. This is documented in the link above.
4. If you don't want to give feedback, you can permanently turn it off for yourself by setting CLAUDE_CODE_DISABLE_FEEDBACK_SURVEY=1 in your env or settings.json file. (docs: https://code.claude.com/docs/en/settings)
5. To permanently turn off the feedback survey for your whole company, set CLAUDE_CODE_DISABLE_FEEDBACK_SURVEY=1 in the settings.json checked into your codebase, or in your enterprise-managed settings.json (docs: https://code.claude.com/docs/en/settings)
6. You can also opt out of both telemetry + survey by setting CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1, or you can more granularly opt out with DISABLE_ERROR_REPORTING=1, DISABLE_TELEMETRY=1, etc. (also documented in the settings docs)
Security and privacy are very important, and we spend a lot of time getting these right. You have full control over data usage, telemetry, and training, and these are configurable for yourself, for your codebase, and for all your employees. We offer all of these options out of the box, so you can choose the mechanism that makes the most sense for you.
If there is a setting or control that is missing, or if anything is unclear from the docs, please tell us!
All this hassle is because you people dont just use Deepseek paid API + Cline. 128.0k context window, the joy of refactoring gigantic applications by paying only $0.05 etc. An average 128.0k context api call costs $0.0008-$0.005.
15 comments
[ 28.5 ms ] story [ 28.3 ms ] threadGet it in an air-tight legal agreement with some kind of audit provision, actual enforcement and penalties, or don't give out data you care about.
Stop asking me about accepting cookies.
Stop asking me to subscribe to your email list
Stop asking me to review my last purchase.
Stop telling me I need to subscribe to view this content.
Stop asking for my phone number
Stop asking for my income
If I want to do or tell you any of these things I will initiate that myself.
Per recent comment from Anthropic at https://github.com/anthropics/claude-code/issues/8036#issuec...
It's kind of shocking how some people just don't get how insanely insulting it is for an application to constantly ask for feedback.
It should at least wait until I've finished my journey and parked up. Not that I'm going to bother giving ad hoc feedback then either.
1. We use session quality feedback as a signal to make sure Claude Code users are having a good time. It's helpful data for us to more quickly spot & prevent incidents like https://www.anthropic.com/engineering/a-postmortem-of-three-.... There was a bug where we were showing the survey too often, which is now fixed (it was annoying and a misconfiguration on our part).
2. Giving session quality feedback is totally optional. When you provide feedback, we just collect your numerical rating and some metadata (like OS, terminal, etc.). Giving feedback doesn't cause us to log your conversation, code, or anything like that. (docs: https://code.claude.com/docs/en/data-usage#session-quality-s...)
3. We don't train on quality feedback data. This is documented in the link above.
4. If you don't want to give feedback, you can permanently turn it off for yourself by setting CLAUDE_CODE_DISABLE_FEEDBACK_SURVEY=1 in your env or settings.json file. (docs: https://code.claude.com/docs/en/settings)
5. To permanently turn off the feedback survey for your whole company, set CLAUDE_CODE_DISABLE_FEEDBACK_SURVEY=1 in the settings.json checked into your codebase, or in your enterprise-managed settings.json (docs: https://code.claude.com/docs/en/settings)
6. You can also opt out of both telemetry + survey by setting CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1, or you can more granularly opt out with DISABLE_ERROR_REPORTING=1, DISABLE_TELEMETRY=1, etc. (also documented in the settings docs)
Security and privacy are very important, and we spend a lot of time getting these right. You have full control over data usage, telemetry, and training, and these are configurable for yourself, for your codebase, and for all your employees. We offer all of these options out of the box, so you can choose the mechanism that makes the most sense for you.
If there is a setting or control that is missing, or if anything is unclear from the docs, please tell us!