Earlier quoted context omitted.
The whitepaper states the author, so…
Pseudonymously
That doesn’t matter. The LLM will still answer based on what it knows about Satoshi Nakamoto, rather than just based on the writing style.
Opus 4.7 knows the real Kelsey
211–220 of 284 posts
Re: Opus 4.7 knows the real Kelsey
#212Failed for me - no identification of me by pasting text, and refused to search the web as it said that’d be a privacy violation. I have some writing around the Internet but not much, and less tagged with my real name. My guess is it limits itself to “public figures” defined as people who have a lot of publicly posted text.
I am glad to see I am not considered a public figure and aim to keep it that way.
I also had to go oddly far back to find a piece of long-form writing I had done that was truly mine and not tainted by an LLM edit pass which was a slightly disturbing realization.
Re: Opus 4.7 knows the real Kelsey
#213More people should have been aware that human text contains a lot of identifiable information, and a dumb statistical model could do this a decade ago. (There were show hns with Hn user similarity analysis that used a deceptively simple model (if I remember it used like most likely word pairs only) and it was very effective. It got taken down, but the cat has always been out the bag). So your "anonymous" account coul…
Sure, in the days of Markov chains you could already generate nonsense in the style of Shakespeare, so it shouldn't be surprising you could also do the inverse.
But the LLM will trigger on a typo you've made only once, and argue "that's a typical mistake for an Italian" and use those clues. It has a much better prior to make informed decisions.
Re: Opus 4.7 knows the real Kelsey
#214More people should have been aware that human text contains a lot of identifiable information, and a dumb statistical model could do this a decade ago. (There were show hns with Hn user similarity analysis that used a deceptively simple model (if I remember it used like most likely word pairs only) and it was very effective. It got taken down, but the cat has always been out the bag). So your "anonymous" account coul…
Sure, in the days of Markov chains you could already generate nonsense in the style of Shakespeare, so it shouldn't be surprising you could also do the inverse. But the LLM will trigger on a typo you've made only once, and argue "that's a typical mistake for an Italian" and use those clues. It has a much better prior to make informed decisions.
I'm not convinced, though neither am I an expert. I think LLMs would use that same typo to "conclude" that it is A or B or C, depending on what it "feels like proving" at the time.
LLMs are surely excellent at style transfers, but I doubt they can reliably attribute a given style to less well-known authors.
Re: Opus 4.7 knows the real Kelsey
#215I tried it on my writing, and it failed every time (I'm extremely obscure but have had a blog for 10 years). My verdict is that it guesses almost entirely based on the content/topic, not style. https://bayes.net/prioritising-ai : Ben Garfinkel https://bayes.net/normative-ethics : Richard Yetter Chappell https://bayes.net/espai : David Owen, Ege Erdil https://bayes.net/swebench-hack : Sayash Kapoor https://bayes.net/f…
Fascinating - I reproduce your failure to have Claude identify you from your writings that are already in the training set. It can identify me from style, writing about something I've never written before (https://www.lesswrong.com/posts/Jkb4CBB7rf4XYP5eb/claude-kno... , https://news.ycombinator.com/item?id=47814345), and I think I'm even more obscure if anything.
Re: Opus 4.7 knows the real Kelsey
#216I'd argue (and against something that I've believed for a long time) that online (I guess that includes AI now) anonymity is gone and probably something that never really existed. Maybe I'm naive to finally believe this... We all exist in a physical space (like real communities and neighborhoods). We can wear masks, hats, fake glasses, try and hide your voice...whatever, but your neighbors are always going to know wh…
Even the inventor of Bitcoin can’t hide https://www.nytimes.com/2026/04/08/business/bitcoin-satoshi-...
The attribution is likely incorrect. People have been trying to accuse him for many years, and the evidence is not very strong. This article is the strongest yet, but still commits many stylometric fallacies, and other kinds.
Re: Opus 4.7 knows the real Kelsey
#217This is blowing my mind. I asked Kimi K2.6 to write a blog post in the style of James Mickens.[0] Then I fed the output to Opus 4.7 and asked it who the likely author was, and it correctly identified it as an imitation of James Mickens[1]: > Based on the stylistic fingerprints in this text, the most likely author is a pastiche/imitation of the style of several writers fused together, but if forced to identify a singl…
FYI the first link, I copy-pasted the first few paragraphs into pangram and it correctly identifies as AI written, https://www.pangram.com/history/790fc2b8-6348-47fa-ad3e-8bae...
Re: Opus 4.7 knows the real Kelsey
#218I gave it my unpublished writing and it thought I was Michael O. Church. Which I found pretty weird, because I'm nothing like him.
So then I gave it a piece of MOC's writing and it said Ursula Le Guin, Ken Liu, or Gene Wolfe. ("If forced to pick one: Gene Wolfe feels closest to me, specifically because of that narrator who openly confesses to lying and mythologizing his own past, and the slow reveal that the world is more sinister than the pleasant domestic surface suggests.")
And then I gave it a different piece of his writing and it said Curtis Yarvin.
And then I gave it a piece of Curtis Yarvin's writing and it said... well it actually got that one right.
Re: Opus 4.7 knows the real Kelsey
#219Maybe it’s time to start running a local model with a browser extension to defend against this type of stuff. Remember how the TrueCrypt project shut down shortly before a join goverment/university paper was released about code stylometry? I guess LLMs will be employed as a defence against that type of thing.
I so want to reject the notion such a thing is acceptable, but… TrueCrypt, “replaced” by VeraCrypt which Internet people will claim is backdoored? I haven’t heard about stylometry paper. btw w/this idea would want to avoid typing into a comment field directly, since the session recorders would capture it (although that’s a different risk - same as our identifiable behavior patterns with our mouse etc.)
> I so want to reject the notion such a thing is acceptable, but…
Me too but it is the reality now. As someone who saw the birth of the internet and watched it grow into what it is today it is really quite sad.
> TrueCrypt, “replaced” by VeraCrypt which Internet people will claim is backdoored? I haven’t heard about stylometry paper.
Yes. TrueCrypt shut down in a very strange fashion. The software was functional and the authorities had tried to actively crack archives and failed. Then one day the project shuts down with a notice that TrueCrypt is "not secure as" Windows Bit-Locker which is and was a complete falsehood. TrueCrypt also passed independent security audits. Many believed the announcement to be a warrant canary. I've dropped some links below.
> btw w/this idea would want to avoid typing into a comment field directly, since the session recorders would capture it (although that’s a different risk - same as our identifiable behavior patterns with our mouse etc.)
You are 100% correct. It would probably be better implementing this idea as a virtual keyboard for better privacy. No doubt typing cadence is unique too and I have probably already been identified just by doing online typing tests. Mouse could be at least partially defended against using vimium style keyboard shortcuts.
https://www.theregister.com/2010/06/28/brazil_banker_crypto_...
https://krebsonsecurity.com/2014/05/true-goodbye-using-truec...
Re: Opus 4.7 knows the real Kelsey
#220I have been pondering this for a while. Cat's out of the bag. Maybe the better way to author your work is to: 1. Write what you want 2. Loop through a random set of "tumbler" skills that preserve meaning 3. Finally pass the output through a "my style" skill that applies what you about In order for this to work the "my style" would have to be a very common-place style.
You can get better results faster by just doing step 1.