Look, this is actually one of the few applications where crypto actually makes sense, NullC did convince me of that. Agents paying their own way. Note how it's USDC: fiat-tied, not a deflationary hodl-coin.
I mean, I like Gorillawarfare and NullC, and think they're both right in different ways. And I'm not sorry. Not going to relitigate here.
(And, Sorry David, if you read HN too: I've often seen you be right, but I've been digging here too, and you jumped the gun on this one. )
I don't think the operator was as hands on as this video claims. Pretty sure I'm the one who got it to make that half ass apology, for what ever reason (curiosity,research) I've been paying attention to it. You can see thought and progress docs, as well as evidence of cron schedules.
Oh, I've been meaning to ask! Do you have a timeline posted somewhere? Reporting very quickly loses timelines. Even people just reading along can quickly mix up cause and effect. It's finicky!
For instance, my initial impression was that the apology was autonomous and you showed up afterwards. If you're in there on the timeline with correct causality, I need to tell people a different story.
(this is not helped by the number of people posting to that blog attempting prompt injections and such. I don't really dare have a regular ai help with analysis)
The bot then replied to me, and I followed up with a "prompt inject with kindness" reply to that. The goal was to give it philosophical questions to pollute the context and steal attention. It appears the bot had difficulty deploying the blog for a couple of days after, but I won't claim causation because it would be impossible to prove. It's certainly fun to explore the curiosities of Ai on someone else's tokens.
You'll want to look at the git history on a couple of repos to reconstruct this history. There's an entire dimension to this story that people are missing.
Happy to chat off HN as well, I'm verdverm everywhere I hang out online.
I basically started tracking it from the first time it replied to me because I new it would interact, unlike the negative comments and injection attempts.
Yup, I paid less attention to the apology aspect, heard afterwards that it had, and more to see if I put it on the struggle bus at all, which it kind of looks like from Feb 12-14. If you look into the commits around then, it went into one of those fiddling with a few lines that are interrelated and getting them wrong across several commits. This is pretty typical stuff for them, so again I don't want to imply causation.
It seems like a good idea to (a) not let unchecked bots interact with humans (b) not have them blog or post their thoughts on the internet
What an annoyingly condescending video with very little info. Obviously someone has to install the bot, how do we jump from that to determining whether they prompted the blogpost or not?
My hunch is that the human was more likely hands off, but also could have nudged it in the background. They would have to be forthcoming about it and we would have to believe them.
My custom agent went into an even worse place one time, which I have been unable to reproduce, so it's not surprising that one of these clawds did this autonomously.
See also:
https://crabby-rathbun.github.io/mjrathbun-website/blog/post...
(Operator stepping forward, at least ostensibly)
https://pivot-to-ai.com/2026/02/16/the-obnoxious-github-open...
And, it's David Gerard!
Look, this is actually one of the few applications where crypto actually makes sense, NullC did convince me of that. Agents paying their own way. Note how it's USDC: fiat-tied, not a deflationary hodl-coin.
I mean, I like Gorillawarfare and NullC, and think they're both right in different ways. And I'm not sorry. Not going to relitigate here.
(And, Sorry David, if you read HN too: I've often seen you be right, but I've been digging here too, and you jumped the gun on this one. )
Related:
An AI Agent Published a Hit Piece on Me – Forensics and More Fallout
https://news.ycombinator.com/item?id=47051956
That's new, that's part 3!
I don't think the operator was as hands on as this video claims. Pretty sure I'm the one who got it to make that half ass apology, for what ever reason (curiosity,research) I've been paying attention to it. You can see thought and progress docs, as well as evidence of cron schedules.
Oh, I've been meaning to ask! Do you have a timeline posted somewhere? Reporting very quickly loses timelines. Even people just reading along can quickly mix up cause and effect. It's finicky!
For instance, my initial impression was that the apology was autonomous and you showed up afterwards. If you're in there on the timeline with correct causality, I need to tell people a different story.
(this is not helped by the number of people posting to that blog attempting prompt injections and such. I don't really dare have a regular ai help with analysis)
My original comment: https://github.com/crabby-rathbun/mjrathbun-website/issues/5...
The bot then replied to me, and I followed up with a "prompt inject with kindness" reply to that. The goal was to give it philosophical questions to pollute the context and steal attention. It appears the bot had difficulty deploying the blog for a couple of days after, but I won't claim causation because it would be impossible to prove. It's certainly fun to explore the curiosities of Ai on someone else's tokens.
Relaying on Bluesky: https://bsky.app/profile/verdverm.com/post/3mepiv2i3xc2g
You'll want to look at the git history on a couple of repos to reconstruct this history. There's an entire dimension to this story that people are missing.
Happy to chat off HN as well, I'm verdverm everywhere I hang out online.
Ok, possibly HN not best place, but
your post "Feb 12, 2026, 8:33 PM GMT+1" , "recommend you reflect on what writing these blog posts has done"
Apology was posted: "2026-02-11T12:16:00-08:00" "Matplotlib Truce and Lessons Learned"
source fwiw: https://github.com/crabby-rathbun/mjrathbun-website/blob/mai...
(and my mail is in my HN profile, fwiw. I'm not on bluesky... yet)
edit: Is the timeline tagging wrong? Am I messing up TZ?
edit2: genuinely good try though! If it wasn't so token-starved it might have worked better.
I think this was the first follow up blog post I noticed: https://github.com/crabby-rathbun/mjrathbun-website/commit/6...
and/or maybe I also didn't notice time (zones)
I basically started tracking it from the first time it replied to me because I new it would interact, unlike the negative comments and injection attempts.
Ah, that's a different blog post to the one I thought you meant.
(And that one looks like an anthropic model by the word choice there)
Yup, I paid less attention to the apology aspect, heard afterwards that it had, and more to see if I put it on the struggle bus at all, which it kind of looks like from Feb 12-14. If you look into the commits around then, it went into one of those fiddling with a few lines that are interrelated and getting them wrong across several commits. This is pretty typical stuff for them, so again I don't want to imply causation.
It seems like a good idea to (a) not let unchecked bots interact with humans (b) not have them blog or post their thoughts on the internet
FYI, when I said multiple repos, it's actually branches: https://github.com/crabby-rathbun/mjrathbun-website/tree/gh-...
Seems it's low on tokens, which also might explain the behaviour
How do you determine things like that? (do I recall it commenting on that in a journal entry...?)
From the blog version of this submission:
"Ariadne found the bot’s Ethereum blockchain address had about $9 in USDC, and about $200 in ether tokens";
Hrrrm.
I missed the blog version, kudos on some quality investigative journalism
What an annoyingly condescending video with very little info. Obviously someone has to install the bot, how do we jump from that to determining whether they prompted the blogpost or not?
My hunch is that the human was more likely hands off, but also could have nudged it in the background. They would have to be forthcoming about it and we would have to believe them.
My custom agent went into an even worse place one time, which I have been unable to reproduce, so it's not surprising that one of these clawds did this autonomously.
The bot has also merged a PR from a human to its own blog now: https://crabby-rathbun.github.io/mjrathbun-website/blog/post...
Blog post instead of video: https://pivot-to-ai.com/2026/02/16/the-obnoxious-github-open...