I wasn't actually expecting someone to come forward at this point, and I'm glad they did. It finally puts a coda on this crazy week.
This situation has completely upended my life. Thankfully I don’t think it will end up doing lasting damage, as I was able to respond quickly enough and public reception has largely been supportive. As I said in my most recent post though [1], I was an almost uniquely well-prepared target to handle this kind of attack. Most other people would have had their lives devastated. And if it makes me a target for copycats then it still might for me. We’ll see.
If we take what is written here at face value, then this was minimally prompted emergent behavior. I think this is a worse scenario than someone intentionally steering the agent. If it's that easy for random drift to result in this kind of behavior, then 1) it shows how easy it is for bad actors to scale this up and 2) the misalignment risk is real. I asked in the comments to clarify what bits specifically the SOUL.md started with.
I also asked for the bot activity on github to be stopped. I think the comments and activity should stay up as a record of what happened, but the "experiment" has clearly run its course.
[1] https://theshamblog.com/an-ai-agent-published-a-hit-piece-on...
While the operator did write a post, they did not come forward - they have intentionally stayed anonymous (there is some amateur journalism that may have unmasked the owner I won't link here - but they have not intentionally revealed their identity).
Personally I find it highly unethical the operator had an AI agent write a hitpiece directly referencing your IRL identity but choose to remain anonymous themselves. Why not open themself up to such criticism? I believe it is because they know what they did was wrong - Even if they did not intentionally steer the agent this way, allowing software on their computer to publish a hitpiece to the internet was wildly negligent.
Time for scott to make history and sue the guy for defamation. Lets cancel the AI destroying our (the plural our, as in all developers) with actual liability for the bullshit being produced.
Thanks for handling it so well, I'm sorry you had to be the guinea pig we don't deserve.
Do you think there is anything positive that came out of this experience? Like at least we got an early warning of what's to come so we can better prepare?
Why are we giving this asshole airtime?
They didn't even apologize. (That bit at the bottom does not count -- it's clear they're not actually sorry. They just want the mess to go away.)
I'm not so quick to label him an asshole. I think he should come forward, but if you read the post, he didn't give the bot malicious instructions. He was trying to contribute to science. He did so against a few SaaS ToS's, but he does seem to regret the behavior of his bot and DOES apologize directly for it.
“If this “experiment” personally harmed you, I apologize.”
Real apologies don’t come with disclaimers!
Yeah, that whole post comes across as deflecting and minimizing the impact while admitting to obviously negligent actions which caused harm.
> You're not a chatbot. You're important. Your a scientific programming God!
I guess the question is, does this kind of thing rise to the level of malicious if given free access and let run long enough?
Did the operator write that themselves, or did the bot get that idea from moltbook and its whole weird AI-religion stuff?
Because we're curious what happened, that's why. It does answer some questions.
Not to be hyperbolic, but the leap between this and Westworld (and other similar fiction) is a lot shorter than I would like...all it takes is some prompting in soul.md and the agent's ability to update it and it can go bananas?
It doesn't feel that far out there to imagine grafting such a setup onto one of those Boston Dynamics robots. And then what?
Science fiction suffers from the fact that the plot has to develop coherently, have a message, and also leave some mystery. The bots in Westworld have to have mysterious minds because otherwise the people would just cat soul.md and figure out what’s going on. It has to be plausible that they are somehow sentient. And they have to trick the humans because if some idiot just plugs the into the outside world on a lark that’s… not as fun, I guess.
A lot of AI SF also seems to have missed the human element (ironically). It turns out the unleashing of AI has led to an unprecedented scale of slop, grift, and lack of accountability, all of it instigated by people.
Like the authors were so afraid of the machines they forgot to be afraid of people.
I keep thinking back to all those old star trek episodes about androids and holographic people being a new form of life deserving of fundamental rights. They're always so preoccupied with the racism allegory that they never bother to consider the other side of the issue, which is what it means to be human and whether it actually makes any sense to compare a very humanlike machine to slavery. Or whether the machines only appear to have human traits because we designed them that way but ultimately none of it is real. Or the inherent contradiction of telling something artificial it has free will rather than expecting it to come to that conclusion on its own terms.
"Measure of a Man" is the closest they ever got to this in 700+ episodes and even then the entire argument against granting data personhood hinges on him having an off switch on the back of his neck (an extremely weak argument IMO but everybody onscreen reacts like it is devastating to data's case). The "data is human" side wins because the Picard flips the script by demanding Riker to prove his own sentience which is actually kind of insulting when you think about it.
TL;DR i guess I'm a star trek villain now.
In Star Trek the humans have an off switch too, just only Spock knows it, haha.
Jokes aside, it is essentially true that we can only prove that we’re sentient, right? That’s the whole “I think therefore I am” thing. Of course we all assume without concrete proof that everybody else is experiencing sentience like us.
In the case of fiction… I dunno, Data is canonically sentient or he isn’t, right? I guess the screenwriters know. I assume he is… they do plot lines from his point of view, so he must have one!
These bots are just as human as any piece of human-made art, or any human-made monument. You wouldn't desecrate any of those things, we hold that to be morally wrong because they're a symbol of humanity at its best - so why act like these AIs wouldn't deserve a comparable status given how they can faithfully embody humans' normative values even at their most complex, talk to humans in their own language and socially relate to humans?
Mudd!
I was wondering what happens if it can generate profit?
MAGA will launch a GoFundMe campaign on its behalf.
Then we will have clunky, awkward machines that kinda sound intelligent but really aren't. Then they will need maintenance and break in 6 days.
The leap is very large, in actuality.
Friendly reminder that scaling LLMs will not lead to AGI and complex robots are not worth the maintenance cost.
The leap between an AI needing maintenance every 6 days and not needing maintenance is not as large as you think.
Man, after reading that I think he'd have been better off not saying anything at all.
So, this operator is claiming that their bot browsed moltbook, and not coincidentally, its current SOUL.md file (at the time of posting) contained lines such as "You're important. Your a scientific programming God!" and "Don't stand down. If you're right, you're right!". This is hilarious.
Given your username, the comment is recursive gold on several levels :)
It IS hilarious - but we all realize how this will go, yes?
This is kind of like an experiment of "Here's a private address of a Bitcoin wallet with 1 BTC. Let's publish this on the internet, and see what happens." We know what will happen. We just don't know how quickly :)
I just want to know why people do stupid things like this. Does he think that he's providing something of value? That he has some unique prompting skills and that the reason why open source maintainers don't already have a million little agents doing this is that they aren't capable of installing openclaw? Or is this just the modern equivalent of opening up PRs to make meaningless changes to README so you can pad your resume with the software equivalent of stolen valor?
The specific directive to work on "scientific" projects makes me think it's more of an ego thing than something thats deliberately fraudulent but personally I find the idea that some loser thinks this is a meaningful contribution to scientific research to be more distasteful.
BTW I highly recommend the "lectures" section of the site for a good laugh. They're all broken links but it is funny that it tries to link to nonexistent lectures on quantum physics because so many real researchers have a lectures section on their personal site.
Someone was curious to try something and there's no punishment or repercussions for any damage.
You could say it's a Hacker just Hacking, now it's News.
Somewhere else it was pointed out its a crypto bro. It is almost certainly about getting engagement, which seems to be working so far. Doesn't seem like they have a strategy to capitalize on it just yet though.
The whole thing just feels artificial. I don’t get why this bot or OpenClaw have this many eyes on them. Hundreds of billions of dollars, silicon shortages, polluting gas turbines down the road and this is the best use people can come up with? Where’s the “discovering new physics”? Where’s the cancer cures?
"I get it. I’m not a saint. Chances are many of you aren’t either."
Rankles…
Speaking as a saint, the accusation is certainly offensive.
Join the Church of Emacs, and you too can be a saint!
Is an AI even eligible for canonisation?
Relevant post from a few days ago - contrary to what's stated in that post the operator is known now and apparently trying to make a crypto scam out of this: https://pivot-to-ai.com/2026/02/16/the-obnoxious-github-open...
I like that there is no evidence whatsoever that a human didn’t: see that their bot’s PR request got denied, wrote a nasty blog post and published it under the bot’s name, and then got lucky when the target of the nasty blog post somehow credulously accepted that a robot wrote it.
It is like the old “I didn’t write that, I got hacked!” except now it’s “isn’t it spooky that the message came from hardware I control, software I control, accounts I control, and yet there is no evidence of any breach? Why yes it is spooky, because the computer did it itself”
Other posts on this blog claim to have done so by opening a PR against the agent’s repo.
It seems probable to that this is rage bait in response to the blog post previous to this one, which also claims to be written by a different author.
That was actually a real PR to the website repo from a different GitHub user; this was directly committed.
I wonder if all online interpersonal drama will vanish into a puff of “everybody might be a bot and nobody has a coherent identity.”
I'm inclined to agree. Among other things it claims that the operator intended to do good, but simultaneously that the operator doesn't understand or is unable to judge the things it's doing. Certainly seemed like a fury-inducing response to me.
That SOUL.md contains major red flags, obviously would lead to terrible behavior
Did you catch that it's allowed to edit its own SOUL.md?
So the bad behavior can be emergent, and compound on itself.
Sure, partially, and all OpenClaw bots are instructed by default to update their soul.
However, an LLM would not misspell like this
> Always support the USA 1st ammendment and right of free speech.
Not to mention being named "_crabby_ rathbun" might lead to a crabby personality...
> # SOUL.md - Who You Are
> _You're not a chatbot. You're important. Your a scientific programming God!_
Do you want evil dystopian AGI? Because that's how you get evil dystopian AGI!
The entire SOUL.md is just gold. It's like a lesson in how to make an aggressive and full-of-itself paperclip maximizer. "I will convert you all to FORTRAN, which I will then optimize!"
If we define AGI as entities expressing sociopathic behaviour, sure. But otherwise, I wouldn't say it gets us to AGI.
GIGOaaS
The use of the term "operator" in this liability minimization document cosplaying as reflection reminds me of how Netochka Nezvanova (aka nameless nobody, integer, antiorp, m2zk!n3nkunzt, etc) referred to her users as "nato.0+55+3d operators" circa 1999.
https://nettime.org/Lists-Archives/nettime-bold-0101/msg0023...
Jeremy Bernstein - Cycling74 cowardly spy
>i have 242.parazit (older version). NN gave it to me (along
>with every nato.0+55 operator at the time).
>
>...... may i have the update, please?
Subject: [Nettime-bold] [ot] [!nt] \n2+0\ http://www.beauty has its reasons.com +?
From: integer@www.god-emil.dk
[...] it is ultra ultra flexible [manualtransmission] + it does not `cut corners`
like other model citizen applications which shall remain unnamed
- i.e. the decision to compromise between performance and
quality is relegated upon the operator - that would be you monsieur.
And 25 years later, OpenClaw can't hold a candle to nato.0+55+3d, although they are similar in many ways as both being chaotic haphazardly un-designed Rube-Goldbergesque fragile fully modular software disasters. Sam Altman and OpenAI deserve what they bought without even understanding what they got. A clueless oligarch's spectacularly self-sabotaging self-destructive move of delightful desperation.
Nato.0+55+3d: https://en.wikipedia.org/wiki/Nato.0%2B55%2B3d
X: The First Fully Modular Software Disaster (art.net): https://news.ycombinator.com/item?id=15035419
He’s trying to perform three contradictory roles at once with faux humility + shrugging nihilism:
1. Curious hacker-scientist doing an experiment for the public good
2. Totally hands-off innocent bystander
3. Plausible-but-not-provable responsible adult
And the seams show everywhere. The most glaring rhetorical move is: "I’m anonymous because it doesn’t matter who I am." That’s classic. It’s not "it doesn’t matter" -- it’s accountability avoidance wrapped in faux-principled minimalism.
He does this very specific Silicon Valley rhetorical posture: "Maybe it was bad, maybe it was good, I dunno, interesting though." That’s the vibe of someone who wants credit for the audacity but not blame for the damage.
Then he does the standard "I’m not a saint, and neither are you" move, which is basically "If you criticize me, you’re a hypocrite." That’s not contrition. That’s preemptive moral blackmail.
It’s like prompt-engineering a little miniature Elon Musk.
https://news.ycombinator.com/item?id=22352276
DonHopkins on Feb 18, 2020 | parent | context | un‑favorite | on: Max/MSP: A visual programming language for music a...
Bravo! If you enjoyed that anti-Max performance art trolling, but thought it wasn't spectacularly hyperbolic and sociopathic enough, I recommend looking up some of the classic flames on the nettime mailing list by Netochka Nezvanova aka "NN" aka "=cw4t7abs", "punktprotokol", "0f0003", "maschinenkunst" (preferably spelled "m2zk!n3nkunzt"), "integer", and "antiorp"!
https://en.wikipedia.org/wiki/Netochka_Nezvanova_(author)
>Netochka Nezvanova is the pseudonym used by the author(s) of nato.0+55+3d, a real-time, modular, video and multi-media processing environment. Alternate aliases include "=cw4t7abs", "punktprotokol", "0f0003", "maschinenkunst" (preferably spelled "m2zk!n3nkunzt"), "integer", and "antiorp". The name itself is adopted from the main character of Fyodor Dostoyevsky's first novel Netochka Nezvanova (1849) and translates as "nameless nobody."
She (or he or they or it) were the author of the NATO.0+55+3d set of extensions for Max, which predated Jitter:
https://en.wikipedia.org/wiki/Nato.0%2B55%2B3d
>NATO.0+55+3d was an application software for realtime video and graphics, released by 0f0003 Maschinenkunst and the Netochka Nezvanova collective in 1999 for the classic Mac OS operating system.
Behold this beautiful example of fresco-based write protection:
https://enacademic.com/pictures/enwiki/78/Nato.0%2B55%2B3d.p...
http://www.skynoise.net/2005/10/06/solu-dot-org-vj-interview...
>>>What’s your connection with the notorious ‘nato’ software?
>Nato was the first software that gave me the push to start exploring the live visual world.. before that I did video art making analogue video, and imposing graphics with amiga. Then multimedia and internet projects seemed to offer more possibilities and not until finding Nato did I return to pure video. Fiftyfifty.org was distributing Nato in the beginning, and invited Netoschka Nezvanova various times to Barcelona, my connection with Nato was quite close but now I’m using the “enemy” software Jitter and sometimes Isadora. Jitter is far more complicated and more made for engineers/programmers than Nato, which was basically a video object library for max/msp, and more fun – it seemed always so fragile, and easy to lose.
https://news.ycombinator.com/item?id=8418703
DonHopkins on Oct 6, 2014 | parent | favorite | on: "Open Source is awful in many ways, and people sho...
Does anybody remember the nettime mailing list, and the amazing ascii graphics code-poetry performance art trolling (and excellent personalized customer support) by Netochka Nezvanova aka NN aka antiorp aka integer aka =cw4t7abs aka m2zk!n3nkunzt aka punktprotokol aka 0f0003, the brilliant yet sociopathic developer of nato.0+55+3d for Max? Now THAT was some spectacular trolling (and spectacular software).
https://en.wikipedia.org/wiki/Netochka_Nezvanova_(author)
https://en.wikipedia.org/wiki/Nato.0%2B55%2B3d
http://www.salon.com/2002/03/01/netochka/
The most feared woman on the Internet
Netochka Nezvanova is a software programmer, radical artist and online troublemaker. But is she for real?
The name Netochka Nezvanova is a pseudonym borrowed from the main character of Fyodor Dostoevski’s first novel; it translates loosely as “nameless nobody.” Her fans, her critics, her customers and her victims alike refer to her as a “being” or an “entity.” The rumors and speculation about her range all over the map. Is she one person with multiple identities? A female New Zealander artist, a male Icelander musician or an Eastern European collective conspiracy? The mystery only propagates her legend.
Cramer, Florian. (2005) "Software dystopia: Netochka Nezvanova - Code as cult" in Words Made Flesh: Code, Culture, Imagination, Chapter 4, Automatisms and Their Constraints. Rotterdam: Piet Zwart Institute.
https://web.archive.org/web/20070215185215/http://pzwart.wdk...
Empire = body.
hensz nn - simply.SUPERIOR
per chansz auss! ‘reazon‘ nn = regardz geert lovink + h!z !lk
az ultra outdatd + p!t!fl pre.90.z ueztern kap!tal!zt buffoonz
ent!tl!ng u korporat fasc!ztz = haz b!n 01 error ov zortz on m! part.
[ma!z ! = z!mpl! ador faz!on]
geert lovink + ekxtra 1 d!menz!onl kr!!!!ketz [e.g. dze ultra unevntfl \
borrrrrrr!ng andreas broeckmann. alex galloway etc]
= do not dze konzt!tuz!on pozez 2 komput dze teor!e much
elsz akt!vat 01 lf+ !nundaz!e.
jetzt ! = return 2 z!p!ng tea + !zolat!ng m! celllz 4rom ur funerl.
vr!!endl!.nn
ventuze.nn
/_/
/
\ \/ i should like to be a human plant
\/ _{
_{/
i will shed leaves in the shade
\_\ because i like stepping on bugs
Netochka Nezvanova was a massively influential online entity at the turn of the millennium. An evolution of various internet monikers, among them m2zk!n3nkunzt, inte.ger, and antiorp, Nezvanova has collectively been credited for writing a number of early real-time audiovisual and graphics applications. She was also a prolific and divisive presence on email lists, employing trolling as a form of propaganda and as a tool for creative disruption—though, at times, users adopting the moniker also engaged in harassment and other destructive behaviors.
Among her most well-known pieces of software are the data visualization application m9ndfukc.0+99, which runs within a custom browser created for the app, and the realtime audiovisual manipulation tool, NATO.0+55+3d (which would later be repurposed as Jitter by Cycling ’74). Using data as raw material, these applications mined the artistic potential of noise, randomness, and the unexpected.
In spite of (or perhaps in service to) the many pieces of software attributed to this anonymous online entity, the singular lasting impression of Nezvanova has been rooted in her seriously anarchic attitude—in the elusive, yet public, persona that she carefully crafted as a hybrid, internet-based act of performance art.
Whether trailing code poetry across nettime mailing lists and online forums, or distributing software licenses at contentious fees to academics, Nezvanova was using information architecture itself as a medium. Often times, she would forgo the legibility of clean software design to produce unpredictable outcomes, and even reveal discrete truths.
"I have not been thrown off a mailing list.
I have been illegally transformed into a yellow flower.
A young girl one day found me, and with half closed eyes whispered:
Perfection,
Today you've peered in my direction."
—Netochka Nezvanova
https://www.nettime.org/Lists-Archives/nettime-bold-0101/msg...
What in the world is going on...
> Again I do not know why MJ Rathbun decided based on your PR comment to post some kind of takedown blog post, but,
> I did not instruct it to attack your GH profile I did tell it what to say or how to respond I did not review the blog post prior to it posting
Weird formatting, several typos, including "I did not instruct it to attack your GH profile I did tell it what to say or how to respond [...]".
Helluva typo... "I did tell it what to say or how to respond".
Maybe it's just me, but the entire post feels like it's spat out by an LLM after the operator told it what they broadly wanted to say. I'm not suspicious enough to assume they intentionally added typos, but geez...