Being anonymous is getting harder and harder
Being anonymous is getting harder and harder
Regain control of your privacy with Proton (and enjoy their Black Friday / Cyber Week deals while they last!): VPN: https://protonvpn.com/blackfriday Mail: https://proton.me/mail/black-friday Grab a brand new laptop or desktop running Linux: https://www.tuxedocomputers.com/en# 👏 SUPPORT THE CHANNEL...
You're viewing a single thread.
It's actually worse.
The video focuses on how you're leaking personal info all the time through the software that you use and the connections that you make, and ways to mitigate it.
However, have you guys heard about forensic linguistics? That's how the Unabomber was caught. The way that you use your language(s) is pretty unique to yourself, and can be used to uncover your identity. This was done manually by two guys, Fitzgerald and Shuy; they were basically identifying patterns in how Unabomber wrote to narrow down the suspects further and further, until they hit the right guy.
Now, let's talk about large "language" models, like Gemini or ChatGPT. Frankly, I believe that people who think that LLMs are "intelligent" or "comprehend language" themselves lack intelligence and language comprehension. But they were made to find and match patterns in written text, and rather good at it.
Are you getting the picture? What Fitzgerald and Shuy did manually 30 years ago can be automated now. And it gets worse, note how those LLMs "happen" to be developed by companies that you can't trust to die properly (Google, Amazon, Facebook, Apple, Microsoft and its vassal OpenAI).
So, while the video offers some solid advice regarding privacy, sadly it is not enough. If you're in some deep shit, and privacy is a life-or-death matter for you, I strongly advise you be always mindful of what and how you write.
And, for the rest of us: fighting individually for our right to privacy is not enough. We need to assemble and organise ourselves, to fight on legal grounds against those who are trying to kill it. You either fight for your rights or you lose them.
Just my two cents. I apologise as this is just side-related to the video, but I couldn't help it.
Conversely, you can now have your manifesto written by a locally run LLM.
“Revise generically: [manifesto]” (certainly not the best prompt)
Folks seem to like Ollama per HackerNews threads: in a coding context here:
Not using Codestral (yet) but check out Continue.dev[1] with Ollama[2] running llama3:latest and starcoder2:3b. It gives you a locally running chat and edit via llama3 and autocomplete via starcoder2.
It's not perfect but it's getting better and better.
Please no unabombing though
Oh wow he wrote a 35k word manifesto… feel like that’s so rare you’d still stand a solid chance at being identified somehow.
You could, but even then you need to put some thought on how to prompt and review/edit the output.
I've noticed from usage that LLMs are extremely prone to repeat verbatim words and expressions from the prompt. So if you ask something like "explain why civilisation is bad from the point of view of a cool-headed logician", you're likely outing yourself already.
A lot of the times the output will have "good enough" synonyms. That you could replace with more accurate words... and then you're outing yourself already. Or simply how you fix it so it sounds like a person instead of a chatbot, we all have writing quirks and you might end leaking them into the review.
And more importantly you need to aware that it is an issue, and that you can be tracked based on how and what you write.
While forensic linguistics is pretty cool, the Unabomber was caught because they released his manifesto and his brother’s wife and brother recognized the unusual phrasing such as ‘Eat your cake and have it too’.
If an author has a large amount of known works then it’s not too difficult to identify other writings by that same author. But if the author does not have a large body of writing that is known to come from that individual, then the best we can do is determine an approximate age and geographic location where the Individual grew up, and that’s only when the unidentified writing is large enough, like in the case of the Unabomber where his manifesto was 30k words.
I did simplify the whole thing, as you noticed; but note that his SIL and brother identifying him is another example of the same process, David knew that expressions that Ted used like "cool-headed logicians" were highly unusual, not too unlike what the socio- and forensic linguists did there.
But if the author does not have a large body of writing that is known to come from that individual
Such as a Lemmy or Facebook account? Or any other online account associated with your writing, really; we produce far more text in the internet than ourselves realise.
And while a priori, your different accounts through different websites might look completely disconnected, as you connect two of them as coming from the same person, connecting a third one is easier. And a fourth. So goes on.
A small caveat is that while the corpus is bigger, so is the noise introduced by people from the other side of the world that happen to use the same patterns as the person whom you want to identify. Even then, I believe that the ability to bulk process text to find authorship grew considerably faster than the number of potential matches.
Additional signal is not noise
Is there a best of on Lemmy
Best of misinformation. The Unabomber was caught because the person that recognized the writing was his own family members. What the hell is this revisionist BS?
Hi there! Looks like you linked to a Lemmy community using a URL instead of its name, which doesn't work well for people on different instances. Try fixing it like this: !bestoflemmy@lemmy.world
You can’t eat your cake and have it too.
Let's see how well this quote ages as subscription services become ubiquitous.
You can’t eat your cake and have it too.
And I can't "magically" know what you're referring to, either - given that you're replying to a rather long comment but providing exactly zero context on what specifically you're replying to.
Quotes, use them.
I don't believe they are intelligent now but I do believe we evolved as basically pattern matching machines.
I don't rule out the possibility of future intelligent text generators, but I don't think that large language models will do it.
I don't know enough about the evolution of cognition in human beings to believe that your second statement is wrong or right.
My second statement is certainly not comprehensive and rather offhand. Its just we have a ton of pattern matching behavior thats been optimized. Sometimes you may read things about it in why we see faces in everything and why we like certain sounds such that we have music and such.
Got it - that's a fair view then.
yeah its just belief thought really. Its made me think on it more and its sorta a chicken and egg thing. like the whole basic things with cells. attracted to beneficial things like food and repulsed by dangerous things. so recognition becomes important but maybe just the like dislike is the thing and not the aparatus to identify it. shrug
How do you define intelligent here, if it’s not what LLMs can do? What’s something an “intelligent” text generator can do that an LLM cannot?
I could go extra lengths on the definition of intelligence and blah blah blah, but at the end of the day:
If you genuinely believe that the current LLMs are intelligent, I can't help but question your intelligence, thus not being worth wasting my time with.
You can inform yourself on how LLMs work through multiple pieces of text across the internet, as well through usage.
If you genuinely believe that the current LLMs are intelligent, odds are that you have nothing meaningful to talk about LLMs, and thus I won't waste my time with you.
