How can I block AI spammers from icalendar?

[niccokunzmann]

As soon as @SashankBhamidi opened #1185, @AiGentsy opened a comment:

That is basically without substance and it would be nice to removed.

How can I block AI spam from collective/icalendar? Does that need to done through the organization?

[SashankBhamidi]

This comment appeared immediately after I added the “good first issue” label. It seems to be automatically triggered by that label rather than a genuine review.

Can we enable interaction limits or moderation to prevent this kind of automated spam?

Or

An organization owner can block the specific account so it can’t interact with any org repos.

Path:
Organization → Settings → Moderation → Blocked users

[stevepiercy]

• I reported the user to GitHub as a spammer.

• I blocked the user from the collective GitHub organization.

  

---

Can we enable interaction limits or moderation to prevent this kind of automated spam?

This is easier said than done, as GitHub doesn't have much to offer.

There are some interesting projects, such as Vouch, that could help.

https://github.com/mitchellh/vouch

I'm in favor an implementation that at least notifies project Admins of the suspected spam, or better yet hold it for review in a moderation queue. I don't want the unintended consequence of silently blocking or banning a legitimate user.

Next, I'd consider enabling the interaction limits setting, although that could have the unintended consequence of temporarily blocking legitimate new GitHub users. If we decide to enable this setting, I'd select the lowest one possible, see how that works, and step it up as needed.

https://github.com/collective/icalendar/settings/interaction_limits

Limit to existing users
Users that have recently created their account will be unable to interact with the repository.

[SashankBhamidi]

@stevepiercy I wasn't accusing anyone, I pointed to commit timestamps and asked for honesty. The policy says AI use is fine but contributors need to own it. If someone denies it when asked and the timeline clearly doesn't add up, that's worth calling out. Otherwise what's the point of having the policy?

[stevepiercy]

The accused did not admit it, and in fact denied it, but you, not believing them, doubled down.

Pointing out details that lead to a foregone conclusion is accusatory. Your follow up was aggressive, unwelcome, and doesn't represent what I think should be the spirit of contributing to icalendar. I don't like it at all.

[SashankBhamidi]

I hear you on keeping the tone welcoming. But I asked a straightforward question on the PR, got a denial, and the commit timestamps don't back it up. The contributor reacted with a thumbs up and didn't push back, so I think the message landed fine.

I'm not trying to make anyone feel unwelcome. The AI policy says be honest about it, and that's all I asked for. I'll leave it there.

[stevepiercy]

You're in a position of privilege and power. Interpret a thumbs up to mean whatever you want, Sir. I'm a mere newcomer.

Think about it.

[niccokunzmann]

Thumbs up! We are all such characters :)

[stevepiercy]

GitHub responded to me, and deleted the offending account.

https://github.com/AiGentsy