There's been some interesting threads about stylometry over the years [1]. The top link was quite decent at unmasking HN alt accounts with basic ngram analysis whipped up in one day [2].
I remember this! it got me pretty good. I have a bad habit of generating alts because I forget the password.
It makes me wonder if we could use non-instruct LLMs to slightly alter the wording of text while keeping the meaning the same meaning. Perhaps by using perplexity or some other metric. I don't know, maybe you compare the distance of the "meaning" vectors.
You might also want to have some "style" vector associated with each pseudonym. For example, I might want it to produce british english under a certain pseudonym, and simulate an ESL speaker under another.
Essentially, you would want some way of re-styling text. The basic way to do this would be to run the same sylometry tools the hunter uses and manually make synonym word/phrase subsitutions to lower your similarity.
It's a cat and mouse game, but I think the mouse eventually wins. Consider a program that translates your english writing programmaticially into a low-entropy symbolic form and then translates them back to english in a procedural manner. Basicially you design an intermediary language that cannot contain style. It would be boring to read but it would remove all the style.
Yep, I was thinking the exact same thing. Write something, get your llm to rewrite it to "standard Steve Yegge form" to hide your identity. If we all do the same we are probably safe.
If that's as effective as comments suggest, then it seems indeed unmasking pseudonyms has only been a matter of effort for a while.
I still think people underestimate the power of even minor inconvenience. While you can't just click a button to reveal all pseudonyms of someone (e.g. you need to download some obscure tool or even perform statistical analysis yourself), I think this provides significant (and surprising?) protection for pseudonymous individuals, I'd say even (although to a reduced extent) for sophisticated threats like state-level actors. I hope LLMs continue to be unable to do so for the near future (I just tested and LLMs can't do it with a simple prompt).
Which is why I think privacy safeguards still work quite well even while being technically mostly bypassable.
So probably tools like this should be kept private if possible.
Looks like the site was taken down. I was curious who wrote the most like me, given that I don't have any alt accounts, but I guess weighing privacy over my curiosity is a good thing in the big picture
"Pseudpocalypse"... why not "pseudocalypse" or "pseudoapocalypse"?
As an aside, it's always surprising to see how English speakers split Greek words like "Apocalypse". That is to say, they always split them in the middle of Greek syllables, or just drop letters like "pseud[o][a]pocalypse" and often in a way that ends up sounding clunky and weird even in English.
Can't think of other examples now. Brain going to sleepzzzz....
Let's see, expanded it's "the (or an) apocalypse of pseudonyms." In Greek, this would roughly be "ἀποκάλυψις ὧν ψευδωνῠ́μων." It's not impossible for a genitive like this to become a word, possibly "ἀποκάλυψευδώνῠμοι" (think of "φιλέλλην") -- although it's a little fanciful. There may even be precedent for a construct that puts "ψευδώνῠμοι" first. But ultimately I agree that the word the article offers seems tortured.
It's a speech and attention thing. With the American accent you weigh the pseud part higher than the -onym because it's the downward part of the word. pseudocalypse highlights the -ocalypse on the upswing which detracts from the pseudo- so to maximize traction use both recognition highlights pseud- and -pocalypse.
I wondered something almost the opposite the other day. With consistent interaction with chat bots and AI output, their linguistic ticks are surely bleeding into every day speech ("smoking gun" and other turns of phrase). What if our linguistic output starts to conform?
> Do they, incorrectly, position their adverbial clauses?
Underrated line.
I'd seen the articles making claims about LLM-idiom being unconsciously imitated by humans, and reassured myself that _I_ couldn't possible be so impressionable.
Then during a presentation I was making today, a coworker wrote in chat "If <scubbo> says 'You're absolutely right' one more time, I'm checking his house to see if we're actually talking to an agent".
To achieve this level of total surveillance it would trillions and trillions of dollars to be invested in a massive roll out of data centers. I just dont see it happening.
I wonder if this could be used to unmask satoshi. I remember a piece about applying stylometry to satoshi's writing, but they just compared him to the usual list of suspects (finney, back, etc.)
I've gotten several dozen accounts banned on reddit over the years. I don't remember the names of them all. It would be hilarious to me if someone mined the archives and stitched them all together.
One would think that it could be possible to make a tool that takes some text and "anonymizes" it by making it a little more standard and boring (uniforming punctuation and sentence structure, changing words with some synonyms, etc). Maybe wouldn't make it particularly compelling, but would be valuable for political dissidents and other people with a high threat model.
Sooo, this implies such deep profiling hasn't been in use for a decade for target advertisements. 500M is not that large a number with the amount of traces we leave behind online :)
maybe by the platforms based on the non-publicly-disclosed behaviour (including usage patterns, like rates of opening specific menus), but not by other unprivileged users seeking to publicly out pseudomymous authors.
So anonymity of written speech is toast. We should, however, strive to preserve other forms of anonymity. For example, donations given to political causes should be kept confidential. Let protesters wear masks up to the point where they break the law.
No just use a text-mixer: in goes your text, set parameters to have output match <scapegoat> (or just pick "standard_Neanderthal_3"), out comes text conveying the message you wrote, in the style of your choosing.
Of course that would also strip the attributes that made it your creation.
Which quotes Tao on using deliberate disinformation to preserve anonymity.
> …one additional way to gain more anonymity is through deliberate disinformation. For instance, suppose that one reveals 100 independent bits of information about oneself. Ordinarily, this would cost 100 bits of anonymity (assuming that each bit was a priori equally likely to be true or false), by cutting the number of possibilities down by a factor of 2100; but if 5 of these 100 bits (chosen randomly and not revealed in advance) are deliberately falsified, then the number of possibilities increases again by a factor of (100 choose 5) ~ 226, recovering about 26 bits of anonymity.
Intentionally adding writing "tics", scheduling posts to appear between 2am and 6am in your timezone, or pretending to have a different gender/location/age should help a lot in staying pseudonymous for a while longer.
I was thinking this. Why not just lie? Come up with personas, stack them on other personas, and eventually you're pretending to be a 23 year old Asian man from California who is pretending to be a 50 year old New Zealand carpenter and half the bits are muddled.
What are you writing about? The real person is going to write about what they know, so when the topic of housing comes up, and they pick New York City to discuss zoning restrictions, sure it's possible they're in New Zealand or California, but it makes more sense to assume the user behind the keyboard is in NYC. The question is, for what purposes is the user trying to be unmasked? In the world we live in, the entirely boring purpose is to sell them ads.
Possibly, but it can be very difficult to conceal age or location reliability, especially as your corpus of text grows. And of course that is with just the text, when you start adding in other bits of digital information it gets even more difficult.
The best mitigation a person has is to have an LLM 'flood the zone' with slop based on her writing style.
So she has one comment on the internet admitting she cheated on her taxes, another copping to an axe murder, another revealing she's the one and only D B Cooper.
That gives a new reason that people should read LLM generated articles. For privacy purposes, if writing is run through Claude to add LLM-isms, then it's harder to deanonymize the writer. If Edward Snowden were posting files online, he'd want to post his writing that way too.
Things like it have happened. The Unabomber was identified through his writing idiosyncrasies.
Lots of criminals have been identified by reanalyzing old DNA samples using data and genealogical techniques that weren't possible at the time the samples were left.
This is the kind of thing that is totally technically plausible, and useful to intelligence groups but not the general public, and therefore I would assume it is 100% in production in international intelligence agencies. But nothing we can "prove" without getting shot.
Similar to ultra-accurate multimedia geolocation models.
Huh, I just realized he (she?) was pseudonymous and not Matt Might[0] this whole time. Oops.
Also, I wonder about this analysis in the age of AI slop. I wonder how much that removes the identifying bits, vs how much carries through of the original prompt (e.g. topic and guidance). It's interesting that a pseudonymous blogs might take on very generic Claude-voice, which could be worthwhile if the topics were interesting, but could also just be a completely humanless bot.
AI is clearly the correct answer to this specific problem: Claude can not identify another writer if they use Claude (or another LLM) enough to massage the text.
The problem, as the article correctly identifies, is that this is just the tip of the iceberg. There are tons of other cases (e.g. identifying a car without a license plate by the scratches) that AI is going to enable ... and in those cases we can't just get anonymity back by using an LLM.
> A stronger conjecture is that we’re heading towards a sort of generalized pseudpocalypse. Perhaps, in the future, if you interact with the world through essentially any high-bandwidth channel, then you identify yourself. Say you wear a mask in public and only speak by sub-vocalizing into a voice changer. That’s fine, you’ll still be identified using your body shape, gait, or chemical signature. Or say you don’t like your car being tracked everywhere, so you stop carrying a phone and you somehow convince lawmakers to ban license plates. No problem, your car will still be tracked using tiny scratches or unique pinging sounds from the engine. Or say you don’t like being tracked on the internet, so you lock down your browser profile, buy stuff only with Monero, and connect through a chain of three VPNs. That’s OK. You’ll still be identified through how you wiggle your finger as you scroll down the page. We’re all just too unique, and the information theoretic limit is coming for us.
Forensic research, NSA, Palantir…
Btw 42. Sleep, eat, have sex, have fun, be useful.
As long as you're fine with losing your distinctive voice (which should be taken as table stakes for people who value anonymity to the extent of worrying about this), it's perfectly reasonable to use tools to stymie stylometry. I'm not even suggesting using an LLM (which may or may not be sufficient, it would be difficult to verify either way); it would suffice to have a tool that rewrites your prose (or analyzes it and flags it for manual correction, etc.) if it isn't written in, say, a form where every sentence is primitive subject-verb-object, limited to the 1,000 most common English words, with no contractions, idioms, or exotic punctuation. Yes, this doesn't completely eliminate all possibly identifying bits of entropy, but it would more than suffice for hiding in a crowd ("hiding" in the sense of obviously standing out as someone trying not to be noticed, and as long as you're also careful about your opsec in other ways, like time of posting, etc).
a lot of these comments have no understanding of basic statistics. for example, did you know, at just 30 words of grammatically acceptable text, at least in english, youre already at more possible combinations of phrasings than atoms in the universe. you cant just throw a problem like that at LLMs and expect them to "just work"
folks should also look at burrows delta - i forgot which books but some folks were able to identify a ghost writer by stylometry alone.
ive said this on many threads, you cant just have text output from an llm (regardless of style / "pseudonym") and have it be "unique" because the nature of the transformer model itself is literally present in the output words. it will be detected as llm output every time. it has to be!
for pure anonymity, i suppose then it is an answer... for the actual craft and art style of writing it is not
Eh isn't authorship linking a whole field of study?
Yes it is, called stylometry. Here's a review paper form 2006 [0]. Its an old subject, with literally thousands of papers. Really wished the author had taken a cursory glance at the literature.
One potential solution here I suppose is to make deanonymization or contributing to it a serious crime. I doubt that will happen in most places, though, since it’s often the government that wants to do this to its own citizens.
I don't have the ability to downvote, and I disagree that the downvote should be used, but I disagree with your disagreement about the downvote.
The proposal that seems to infringe on free speech, but mostly I take issue because there just isn't really a precedent for what such a thing would look like. What constitutes doxxing? It is easy to accidentally doxx someone by mentioning some offhand fact in conversation, because you may not know what series of facts can be connected to form a doxx. Typicially, people doxx themselves by accident, and someone else is merely pointing it out. I know I have many times. You live and learn.
There isn't really a right to privacy in the way that there's a right to free speech. Free speech is important to processes of public transparency and justice which I think this would interfere with. But it's also true that justice is blind and doxxing can interfere with a judicial process.
Also, if everyone is going to have access to super-stylometry tools in the future, which is the premise of the article that GP reacts to, it will be fruitless to uphold such a right because anyone can just run doxxyou.exe themselves. There's no need to spread doxx because it can be reproduced individually.
The term doxxing came from a certain hacker culture where it was implied that you connected some real-life identity to a criminal pseudonym. It essentially meant "snitching". So the idea that doxxing itself would be a crime is interesting. It's a reversal of the original meaning.
But it's also true that the public has discovered sybil-suseptible techniques (like swatting) where you can screw with ordinary people's lives by knowing their identity. Which is interesting because we live in a new culture of "share everything online" vs old the hacker ethos of "don't use your real name online". The attackers have become stronger and the defenders weaker.
Let's say I offhandedly mention your email in this comment, for example, like we had an out of band conversation somewhere else. Then, someone takes that email and links it to another forum where they find your phone number. Then another person looks up that phone number in a phone book. Then another person finds your github account from the email associated with your PGP and finds where you work etc. Let's say a comment you wrote 7 months ago vaguely mentions something about where you live, which in combination with the previous information narrows it down to a single place.
At what point does that become doxxing and who is responsible? Pseudonymous people nessisarially, slowly leak small amounts of information about themselves when on the internet in order to engage in communication.
It is only when you collect "enough" bits of information that it becomes threatening doxx.
You could say publishing this collection of facts is the doxx. But the point of the article is that stylometry and AI tools can do this investigation for you. Anyone can trivially assemble the collection themselves.
I think ultimately, it's a matter of personal responsibility. It has to be. If you have perfect opsec, I can't touch you AI tools or not. And if I have perfect opsec and I'm doxxing you, then no laws will be able to catch up to me.
1. https://hn.algolia.com/?q=stylometry
2. https://news.ycombinator.com/item?id=33756141
reply