wongarsu 11 hours ago

The manifesto is somehow devoid of coherent arguments. You are still personally listed as the author of your commits, adding a co-author is not a mechanism for moving all accountability to that co-author. That's not what a co-author is.

I suspect the author of the website forgot to include in their prompt why they actually dislike AI Co-Author attributions. Or maybe they just have really bad colleagues

And that's while there are plenty of good reasons not to list the AI as co-author. I stopped letting Claude include it

  • zethraeus 10 hours ago

    Funny story: I made this because despite having the global no-attribution setting turned on, Claude still regularly tags itself.

    I'm sure it's a context thing — but boy is it annoying.

captn3m0 1 hour ago

When I read the title, I thought it would be for research papers.

vindin 11 hours ago

Broken on mobile, and clearly written by an AI co-author.

This sounds like Claude from a mile away

“It strikes the tool. It keeps the person. Matching runs on both the display name and the email, normalized — so a renamed bot or a numeric noreply address doesn’t slip through. A human is never in the blast radius.”

  • zethraeus 10 hours ago

    Thanks for the mobile heads up!

    The manifesto is me. The website is not — and I'd like to argue that's very non-ironic!

    Here's the making of, linked at the bottom. https://no-ai-coauthors.dev/making-of.html

    • mrcslws 9 hours ago

      I can confirm, this is how Adam talks.

    • brazukadev 7 hours ago

      it is actually very ironic. You can manually commit your code to easily hide you had a co-author.

zethraeus 11 hours ago

AI Co-authors are impressive marketing, but deleterious to dev work.

This manifesto website includes a git commit-msg hook for your favorite hook manager, and a GitHub Action for preventing contributions containing them from landing in your project.

josefritzishere 9 hours ago

AI can't be accountable therefore AI cannot be the author.