We're running out of nice.
People used to post pictures of their dogs. Now a stranger named @TruthHammer1974 will explain why you're holding the leash wrong.
Global niceness is at critical levels. We've formed a committee. Unfortunately, it's a robot.
The list is real. The global niceness statistic is absolutely not.
Global Niceness Index™
CONDITION
PROVISIONAL, OBVIOUSLY
Entirely fictional index. Regrettably real comment sections.
A naughty list.
With receipts.
Submit a post link. NiceBot opens a file. A human reviews the evidence and approves any public notice. The evidence goes in the file. We are apparently at the stage of civilization where this needs to be a product.
Someone leaves a nasty reply.
Under a photo, a milestone, or an otherwise innocent picture of a golden retriever.
NiceBot flags it. You decide.
Clear hate can be flagged automatically. Borderline snark gets a human review. No secret algorithm goblins handing down sentences.
The receipt goes on the list.
Confirmed comments are saved as evidence. People can share or subscribe to lists, protecting more than one comment section.
OFFICIAL RECEIPT · #DEMO-001
CONFIRMED
NiceBot is not here to ban everyone.
Just to make being a jerk slightly inconvenient.
Some people deserve a second chance. And some people, very specifically, need to type an apology while other people can see it.
Check the list.
Curious whether a handle has made the Department's naughty list? Search the official docket. A listing comes with evidence, a human decision, and a path back to nice.
The case files
Looking for @demo_troll? That one's fictional — inspect the sample receipt under How it works.
Report a naughty post.
Spotted someone being unnecessarily cruel? Send the Department a link. You don't have to be the target, and you don't have to identify yourself. Every report is checked against the original post before a human decides whether it belongs in the docket.
File a report
Two fields. No account required. Absolutely no eight-page application.
🔒 No name or email address required. A report is a request for review, not an automatic listing.
What happens next?
- Link checked. We verify the post is real and accessible.
- Human review. Context matters; criticism alone isn't a crime.
- Decision logged. Only confirmed cases join the public docket.
One report per post is enough. Mass-reporting doesn't make the evidence stronger. Reports are anonymous; the Department keeps links, not reporter names.
The path back to nice.
Congratulations. You've been caught acting like the comments section. Fortunately, unlike middle school, NiceBot provides a clear path to rehabilitation.
The required period of public adulthood. Delete the apology early and the clock starts over. NiceBot has no social life. It can wait.
- “Sorry you were offended” is a weather report, not an apology.
- A new insult in your apology restarts the clock. Especially a clever one.
- Screenshots and vanishing stories don't count. The public post needs to remain up.
- Each listing needs its own apology. Yes, even if you are very busy.
Apologize publicly.
On the same platform where it happened. No DM. No smoke signal. A normal, visible post.
Tag @nicebotapp.
The registrar needs to know where to find it. Untagged personal growth is not administratively actionable.
Leave it up for 48 hours.
Once the apology stands for two days, the listing is removed.
Post the photo.
Send us the nonsense.
Your dog's Halloween costume does not require peer review. NiceBot exists so you can share joyful, boring, perfectly normal stuff without inviting a panel of unpaid critics.
You send it. We review it.
Submit a Threads, Instagram, Facebook, X or Reddit post link for review. Nothing is listed or published automatically.
Judgment by actual people.
For gray-area comments, you make the call. The robot has no authority to be personally offended on your behalf.
Share the paperwork.
Subscribe to other people's lists. A confirmed offender shouldn't get an unlimited audition tour of the internet.
A small amount of peace and quiet. Now available in limited beta, subject to the usual administrative nonsense.
Questions from the list.
NiceBot is a real civility project in private testing. This page is the real registry. Its chart is fake, its demo troll is fictional, and its confidence in bureaucracy is sincere.
Is this a real thing or an elaborate bit?
Both, in the grand tradition of the best bad ideas. NiceBot is a real project in private beta. The fictional Global Niceness Index is not a real measurement, however convincing the stamp may look.
Who decides whether a comment is mean?
Obvious high-confidence hate can be flagged automatically. The person who owns the post makes the final decision on borderline comments. A listing should have a saved comment as a receipt.
Can I apologize my way off the list?
Yes. Post a public apology on the same platform, tag @nicebotapp, and leave it visible for 48 hours. No payment, no backstage negotiations, and no “sorry if you were offended.”
Can I pay to get off the list?
No. We conducted a rigorous internal review and determined that cash bribes would make the robot insufferable.
Does the bot read my private messages?
No. NiceBot reviews links submitted through this website. Social inbox intake is not connected.