Stop Advertising in Your Commits

teaearlgraycold

I have to assume these aren’t just ads, but also a critical RLHF avenue for Anthropic. I imagine they scrape these commits from GitHub and compare them against what Claude provided to the user. If the diff that is pushed is different it means the human had to refine the LLM output and that can be fed back in as training data. Presumably semantic search could enable you to find the matching Claude Code session.

lifis

Huh? It's not advertising, it's disclosure that the code was not fully (or at all) written by you.

aleda145

I don't know, it is a useful signal that the person did not think deeply about their code changes, and should be treated as such.

ishan0102

I much prefer this over the alternative where people use AI to code without anyone knowing.

pkamb

Sent from my iPhone

jeroenhd

I want to know when things are slop or not. At least programmers are willing to admit they're generating slop, unlike social media and blog posts. Also, if I publish something online, you don't get to tell me what I can and can't put there (except for reasonable exemptions for hate speech and such, of course). If you don't like repos that tag their slop, go read someone else's code. Feel free to write a filter in your adblocker for the dozen AI tools you usually find.

arikrahman

Same for stars on repositories, it just incentivizes botting for startups.

nikhilpareek13

if disclosure is the goal, git already has the Co-Authored-By trailer convention for this. I add it manually when I want to flag that an AI was meaningfully involved in a commit, and it shows up properly in Github's UI as a co-author. The claude and cursor footers being default-on rather than opt-in is what makes them feel more advertising than disclosure to me.

sestep

I mean, sure, except that many large open-source projects (e.g. Linux [1], Nixpkgs [2], etc) require this as part of their AI policy. Omit attribution in your own projects if you want, but the maintainers of these projects are owed at least that level of transparency for contributions. [1] https://docs.kernel.org/process/coding-assistants.html [2] https://github.com/NixOS/nixpkgs/commit/d18b8f3238abdb2cd878...

kazinator

I don't agree. Who/what wrote the commit is definitely part of the commit. If Git commits formally had a co-authored header, it would go there. As it stands, there is one author and one committer. If something was pair programmed, whether with a human or machine, you need a commit message trailer if you want to show that. Commit message trailers are a formal mechanism in git, supported by tooling; there are git commands to add and remove them. Totally agree about "sent from my fartphone", of course. Disclosing things in the pull request is not enough; pull requests get lost in the sands of time. Years down the road, all that some downstream consumer has is the git history, not any CI-related metadata.

AlfredBarnes

I was gobsmacked when the Amazon share sheet put "or ask Alexa, your shopping assistant for more ideas"

JeremyHerrman

Is AI responsible for the committed code? Should AI be blamed when services go down due to the change? The answer is absolutely not - the developer is responsible whether the code was AI assisted or not, and the dev's name should be attached to it just like any change. The OP is right: these are ads, plain and simple, and it's a dark pattern for these companies to have attribution enabled by default

intentfy

I turned it off so future agents aren't biased in favor/against a piece of work depending on the author.

epistasis

I have to ask Claude to stop doing this about every two days, and usually I don't see it until after a push to a remote repo. So annoying. Just stop, Anthropic, please. And pay attention to the request to stop, instead of silently turning it back on again all the time. Latest thing was linking to the Claude session that generated some of the PR. Put in somewhere that a commit had LLM assistance, fine, but don't spam everybody please, ESPECIALLY in all the icons all over the GitHub interface. Sheesh. It's already obvious that it's coming from an LLM because it's been overdocumented with excessive prose, and the code is overly verbose.

victorkulla

https://fshot.org/techzone/hacker-news.php

knorker

When I read commit history I want to see the reasons. Commit messages are for extra context. It's very useful if it says AI/LLM was used, then I know that there may not actually be a reason for the choice in the commit, so per Chesterton's fence I can then tear down that fence. Now, do I need to know which brand of LLM? No. And fair enough, I'll stop being specific.

fishbacon

I was hoping for a kind of joke. Like saying "idiot" as the last thing in your post and getting "Please don't sign your posts" as a reply.

zetanor

commit 85cd835e5923cddc1882e74354eac8dba6a925c1 (HEAD -> master, origin/master, origin/HEAD) Author: John AuthorDate: Fri May 22 13:25:33 2026 -0000 Merged PR #197 Tired of planning for dinner every day? FoodDrop™ is the premier ready-to-cook meal-by-drone delivery service in the greater Vancouver area and Belize. Get one month of food dropped onto your driveway by FoodDrop™ for only $12.95 when you use this commit's hash as a coupon code! Offer expires Fri May 25 13:25:33 2026 -0000.

benced

I think this is more of a corporate metrics tracking than advertising. Decision makers aren't seeing these ads in commits but they certainly are seeing a report from Anthropic that "75% of your commits last quarter are from Claude code".

p0w3n3d

This is a small "getting used to" technique to let people "be grateful" to the "ai friend"