Substack has partnered with Pangram to enable readers to scan posts, notes and comments for machine-generated text. No AI detector is flawless, but Pangram is pretty much the gold standard.
AI is nothing more than a tool. Like all tools it has its strengths and weaknesses. It is not a substitute for people or judgement. The management people who think so are idiots. But they were always idiots. Overpaid idiots.
What would you say if I told you that the entire Scamdemic global response was the product of the most sophisticated "AI" tool available to the US/UK "national security" apparatus. Managed by idiots who were always idiots, overpaid idiots? Who did and said what their supersmert "tool" told them to do.
Because it had nearly all of the characteristics of an "AI" product that idiots incapable of discerning real intelligence from artificial intelligence like eugyppius describes the prose idiot writers who use LLM to write end up producing. Slop. What if the Scamdemic response was "AI" Slop? And those who make decisions actually are idiots, not just figuratively or rhetorically?
Here’s mine: What if it wasn’t? In fact I’m sure it wasn’t. Fasci didn’t need AI. He just needed his buds to cover for him, which he ensured by holding the money levers. Garden variety corruption on steroids. No AI required.
Fantastical? Perhaps. Definitely not at the highest levels, we'll agree to that. Now we step it down to the next tier of deciders, the "data scientists" and modelers, the Bachelor's, Masters, Doctorates in epidemiology. Do you know what a degree in epidemiology entails today? Some biology, virology, public administration, 'public health,' and sociology, psychology developed as social and behavioral science.
The soft social sciences are also known as 'pseudosciences.' Because while they attempt to implement the systems of scientific review, analysis and discovery, they are inherently unreliable, always have flaws. While they may approximate the "science" to within 80-90% accuracy on their best days, best conditions, they will never achieve 100% accuracy. And they cannot specify individual behaviors or conditions, just that a pretty good approximate of how many total behaviors or conditions can be reliably expected. Real science must produce much more exacting results. Randomized, Controlled Testing. RCT, 'gold-standard' of science is not possible with soft science. Results cannot be reliably replicated down to specific causality. Pseudosciences give pseudoscientists an *idea* of something, some factor that causes an outcome. But results may vary. And often do.
This is the "science" that computer models employ. The same "science" that attempts to model weather. Weather that barely gets day-to-day, hour-to-hour predictions in a specific location correct, much less broad areas, long-term forecasts. Much less global climate changes. Models.
"AI" uses this type of science, pseudosciences, and attempts to make them precise, match the same level of predictability as hard sciences like physics, engineering, biology, chemistry, etc. And uses the same programming biases that their programmers create within their code, assign "credibility" to sources of information according to those programmer biases. And produce models, outcomes, suggestions with that all-too-familiar "AI" authoritative voice that we're beginning to recognize when we see it/hear it. "AI" will make up bald-faced lies and assert them with certitude, confidence, mocking, belittling and demeaning user input that contradicts them.
Hypothetical: You're a researcher at CDC/NIH/Gates Foundation/Rockefeller Foundation/Wellcome Trust/any large 'public health' research operation. Got your hands on the latest and greatest "AI" system in use in 2020 in the rarified air of a declared national security emergency. You come across studies - with "AI" assist, written by eight-year-old's that say standing six feet apart from other people will stop the spread of a superscary virus. All sorts of similar junk science. That AI programming - by actual idiots - assigns a high degree of credibility to. And your models start spitting out that sort of pseudoscience 'slop' combined with the slop of psychology and sociology, applied behaviorism, "nudges" that are believed to be able to manipulate the public to do what is in their own best interests...according to actual idiots feeding idiotic pseudoscience into the latest, greatest, most sophisticated "AI" model found in the highest reaches of national security.
Which is presented just as authoritatively as "AI's" widely used today do, just as dismissively, mockingly, belittling and demeaning of other ideas. Said with 100% certitude, confidence by the research 'scientists,' actual idiots, to their managerial bosses, 'public health' leaders, politicians, media. Who also are actual idiots, mistake that certitude and confidence for The Science™ that cannot possibly be wrong. "AI" said it. And "AI" is supersmert - to actual idiots.
It becomes one giant feedback loop of AI slop. And all the while, those at the top, at the highest levels smugly and knowingly smile at one another, having made use of the idiots who are in the next tier of managers of society. Themselves aware of the predictability and limitations of pseudoscience, that 80% of them would do what the top level knew they would do, while the 20% who didn't could be and were coerced or eliminated from the equation. All game-theoried out. Because, yes, they game theory this stuff. It's what Jeffrey Epstein and Martin Nowack were proficient at, game-theory modeling for humanity, numerous academic works in the field. And they are not alone.
A hypothetical you reject? Is your prerogative. I don't think its all that far-fetched. And I know that the behavioral command and control system that is being built around us, Larry Ellison's "AI cameras will make you behave," digital currency, social credit systems, etc. will be applications of the exact same scenario I am describing. Pseudoscience producing Minority Report-level "Pre-crime" policing, loss of rights, punishment for thoughts and ideas, preventative, so you never commit the crime their pseudoscience AI models say you will without intervention.
Same. Damn. Thing. And the Scamdemic was a Beta test for numerous aspects of the "build back better" (pseudo)scientific world that's in store for us. Unless we do something to change that.
I think the truth is much more prosaic. Fauci & company were told to stop doing GoF work. The made alternative paths to continue the forbidden work. One of these involved WIV. WIV had notoriously poor isolation protocols. An experimental GoF virus escaped WIV and got out into the population. As soon as it was sequenced the cat was out of the bag. To cover up the GoF they contrived the pangolin dodge. Then they moved to how they could make money and political hay on this pandemic.
The political hay was to throw the election by closing down the economy. Standard modeling, which had always over predicted the extent of an epidemic, came in handy for this purpose and was used to that end. The overwhelming majority of politicians were easily convinced and joined the bandwagon. Perhaps this is what you meant by mind virus.
The money angle came from the shots. In order to get the shots on the table and make money they had to suppress alternatives and extend the pandemic. There was a lot of tactical maneuvering, censorship, & nudging engaged in to achieve these ends. Again the politicians were willing, if ignorant accomplices.
None of this required AI which was largely unavailable to everyone involved because it wasn’t mature yet. And it wasn’t necessary.
1) This CFR, particularly the Q&A section shows exactly how they think about GOF/DURC research restrictions. An obstacle to be overcome, put in place by dumb, emotional anti-science rubes:
Biotechnology: The Potential and Perils of Innovation
2) You are mistaken about AI tech in 2020. It was plenty sophisticated to do as I describe. Consumer end that we see today is yesterdays tech found in national security apparatus. Dumbed-down, of course. If your/my eyes see it, can use it, security services saw it, used it years ago. Your assessment of capability is flawed.
3) What you say about what some higher ups in the Scamdemic can be true at the same time as what I wrote can be true; they are not mutually exclusive. There's a whole lot of useful idiots between the top tier and us. Many of whom used all their education and training in public health policy, epidemiology to just follow what they saw other peers do on the news. Monkeys seeing, monkeys doing. And idiot peers who used existent AI models armed with certitude sounded quite authoritative to other idiots.
4) The entire debate over lab leak vs zoonotic escape is a false binary. It was neither of the above. "Viruses" don't exist as portrayed. They are something, but not as described. A trailing indicator, not leading causality. And much evidence exists that "vaccines" shed and cause illness to self and others. All types, LAIV especially, but even mRNA, which Fauci himself admitted. And the 2019-2020 influenza quadrivalent was both LAIV and engineered with a very aggressive component, known to designers to produce harsh symptoms. The 2019-2020 flu "vaccine" caused what was experienced that winter. Used as the predicate to lockdown the world, allow the mass experiment with mRNA platform, that resulted in the most deadly biological weapon attack on mankind, ever. Taken up willingly by an ignorant, trusting public. Conscripted as "Pioneers."
Building a better flu vaccine – one you don’t have to get every year
AI isn’t that powerful, even in the hands of the IC. AI is a stochastic parrot on steroids. Its primary use would have been supporting fake narratives with altered videos and communications. But we already have plenty of tools for that. In fact, Hollywood is a continuing display of the technology. So AI is neither essential nor sufficient nor even a significant player in the events, even if it was as powerful as you claim, which it is not.
Viruses are real. They exist and have existed for a very long time. We can image them on electron microscopes, determine the genetic makeup, identify their proteins, etc.
Please consider as an alternative the possibility of mental virus infection, then re-write your essay, include links or other citations of primary sources, and publish the essay, for your suggestion is very interesting.
Exactly. Like any tool, the user must know how to effectively employ it and when it can be dangerous with misuse. It is essential that all information provided by AI be verified if it is used to make decisions. The user needs to “ trust but verify.”
In my line of work, clients or potential clients approach my firm and ask me to make stuff for them. Engineering. Details not important
Often I get small or solo operators trying to get some product off the ground.
I had this guy who just could not stop using AI. In his product design, emails to me, whatsapp messages, you name it
Eventually I just called him and told him that communicating with me using a machine is extremely insulting and easy to spot, and makes him sound like a retard. He didnt say much but sent a message later. Guess what was in it
This is not meant to promote AI, but I have an important, quite intelligent business associate whose written communications, whether due to an inborn linguistic handicap or from playing too much football in an American High School, were often so baffling, I had to ask for clarification. Then all of a sudden their communicative effectiveness rose to never-suspected heights: and he confessed having delegated his writing to AI. For the purposes for which we communicate, that is a fine solution -- but, I hasten to add, not a universal one.
AI in this case is allowing a smart guy to sound like he is supposed to. In my case it amplified the moron. I told the fellow that simply being ignorant and clueless was a better and more honest approach, and would actually let ppl help him. Advice didnt take
Yes, well, I asked him about it, and he readily admitted having delegated his semantics to AI. But as I said, doing so HAS improved our communications ... just as a lousy car can still get you from A to Z..
Apparently, a high level of communication skill is required to do the job in question, and the referenced person was incompetent for the job when hired and still is incompetent.
My son's girlfriend asked me what I thought the difference was between AI and humans. I off handedly said, it was our errors/mistakes. I have returned to the statement not made with deep thought and realized it was profound for me. How can one have the lessons and triumph of Hansel and Gretel is mistakes in judgment are not made? For heaven's sake, one cannot have Christianity with no error nor its counterpart the power of redemption.
I understand where you are coming from and appreciate your sentiment. However, AI is *not* the arbiter of truth. Far from it. To imagine that whatever question you pose receives the “correct” answer may prove your undoing. I have experienced personally error in AI response, and as Eugyppius seems to allude, careful and thoughtful questioning can even make your AI concede to your points.
The above touches upon my greatest fear—that AI will make the incompetent appear competent. Incompetence being an inability to understand and question the response produced via AI inquiry.
AI might not make many mistakes to the layman of a topic, but ask it something you really know something about and usually it makes either glaring mistakes or gives you an elaborate sounding beginner take. Also, if you can't tell AI prose apart you need to start working on your taste in literature: it can only produce middle of the road fare and uses the same tricks over and over again. Once you spot them, you can always tell
That is a conundrum isn’t it. If one is relatively well versed on the topic at hand, why ask AI for further information? And if you know very little, then how does one decide whether AI is “off the mark” wrt its response? I believe that leads into my comment wrt use by those with “… a thoughtful and experienced mind in general matters of inquiry and proof…”. Alas, that might be only a fraction of a fraction of society.
My personal use is either quick small facts or finding my way into a topic (related terms, authors and so on) or to explore a wild thesis and see if people are on it already (often the case). But as the saying goes: every new tech makes the smart smarter and the dumb dumber.
I absolutely agree, but people are treating it like it is. I read the essays from the Cosmos Institute. A recent one discussed how representatives of the world's great religions were there to restrain AI models. The writer seemed to believe this was a good thing. I responded that the premise made no sense. But it did highlight the problem. As humans create these LLMs, what are the starting points? Is it Bentham's Utilitarianism? That could go very badly. What is interesting is that there seems to be an agreement that there is an objective truth out there. So, ironically, as Plato cannot be taught in universities for its triggering effects, AI may become one of its students.
“…but people are treating it like it is.” That was precisely my point. If you are not especially well grounded in the subject matter of your inquiry, and further, have a thoughtful and experienced mind in general matters of inquiry and proof, you of course will undoubtedly assume what AI responds to your inquiry is “the truth”. I cannot help these days thinking back to my university undergraduate years and the faculty I took courses under. Every word they spoke I assumed was the “revealed truth”. Only after a lifetime of experience and further study did I realize just of naive I was at the time and just how human they (faculty) were wrt to assumptions and bias.
Even wrt relatively mundane inquiries, AI often sufferers from the selection of information fed to it. I have seen two different responses from the same AI product, which were “taught” from differing sources of information. I have also seen such from different AI products to the same inquiries, which may suffer the same problem of differing “training”. Given that human “knowledge” (leave that broadly defined for now) is said to increase exponentially, I don’t see a solution to this problem.
But will human knowledge keep expanding if societies demand that AI make decisions? Everyone points to Brave New World or 1984 as our dark future. I think the movie Wall-E is more scary. The humans as blobs watching media while the robots were the saviors.
That is not the knowledge I describe. So far AI (as my feeble understanding allows) compiles, evaluates (leave that loosely defined for now) and expels/returns information that it is fed. Knowledge that is new, examples: research papers from science, events that occur in society, or simply the daily fluctuation in weather recorded…will continue to feed “the beast”. What I believe you are touching upon is what is termed AGI, or Artificial General Intelligence. As I understand it (caution, my knowledge here is very limited), we are not there yet.
I tend more towards the idiocracy model, with humans watching Ow My Balls while the bots are making their nut going long on Brawndo while telling the government to use it for everything.
Damned if I understood any of that. Rather than too much AI, I suspect there was a surfeit of philosophy and metaphysics in that comment. Whatever happened to, you know, a clear style?
Hah! I play in the sandbox of philosophy because 7 years ago, my then high schooler told me about White Supremacy and asked if I knew who Hegel was. I had no clue and confused him with Hobbes - whose book the Leviathan I read only part way. Thanks to YT, I have my "masters" in philosophy. A friend describes it as gobbledygook. While I appreciate the sentiment those who are creating such systems do not. One tech bro described himself as a Hegel devotee. That made my blood run cold. Seven years ago, it would have meant nothing.
The difference between AI and humans is this: Humans think, and AI imitates thought. But if you were to ask me to define thinking then I'd have to write an essay.
Apparently, the person referenced in Robird's comment was lacking in one or more mental skills or the moral commitment required for the job when hired.
Many years ago, with the advent of email, my mother made the logical observation that; as the quantity of information increased, the quality would necessarily decrease. Obviously there are exceptions to this, but it's ultimately accurate. She and her siblings exchanged a "round-robin" letter, and they would laboriously hand-write or manually type the family news and drop it in the mail. Now we must skim through the random thoughts of millions of people we don't know and try to find some semblance of meaning, often ignorant of the people who should be the closest to us.
I’ve been following Eugy on X and he has been harassing one individual in particular, who is guilty of getting tons of likes and only writes AI slop (I mean the stuff is REALLY bad and meaningless and he gets tons of followers). Eugy keeps posting his AI meter after the posts and it declares the post to be 100% AI, I’ve been laughing my ass off the whole time. The X algorithm and Grok should be able to identify AI and demote its visibility, likewise promote certain accounts who have many insightful things to say without the use of AI. So many platforms are becoming TikTok AI crap, and this is the dumbing down of our future. It’s akin to the educational “experts” who demanded we teach classes based on the lowest IQ in the room, for fear of leaving them behind. Otherwise, X should offer a special version with zero AI and just human interaction, no cheesy videos, so real and proper discussions can take place again.
I personally use AI for medical studies and journals for research on my daughter’s autism and seizures, and it has led to her seizures completely stopping with the use of a different form of B9 (folate). So, until they start blocking/controlling it, I’m sure going to keep using it!
X has honestly gotten so bad, I’m actually bored a lot of the time now. It’s all the same stuff. I think whatever tweaks they are making to incentivize addiction have fallen flat for a lot of people and it broke the spell. I’m curious to see what happens in the next 6 months.
I couldn't agree more about the problem. AI is ruining content platforms of all kinds. A video by a real human about how AI is distorting history was eye-opening; Google searches return answers to historical questions based on AI-generated YouTube videos that are just plain factually wrong. (https://youtu.be/iUfP6amE7UA?si=dzgzWP88_oRJw9nC)
What to do about AI slop is less clear. I disagree with statements like "despite its considerable powers, AI will probably never be a great prose stylist". "Never" is a long time! There's no reason to think that AI won't keep getting better in many ways, including writing content that no AI detector can reliably distinguish from being human-written.
So what's to be done? That's a tough one. Some kind of chain-of-custody verification, where you not only have to prove you're a human, but show your work in creating the content? There's an opportunity for whoever can come up with a good solution.
It's a toughie. No doubt some windbag in Westminster, Wormtongue in Whitehall or Brigand in Brussels will put up Digital ID as the solution. "For the children', of course. They'll use the Swedes as the paradigm - look they've been injected with ID chips already & it's so convenient. . . etc..
Lately I have watched a couple of YouTube videos by Ben Hodges, the retired US general who lives in Frankfurt. His topic is invariably the Ukraine war, and I don't hear much that I disagree with, but in the course of the video he starts conveying a weird kind of body language that doesn't look natural because of its repetitive limitation. It actually induces me to quit watching, because it's creepy; but I suppose it saves him -- the real General -- time.
If you ask an LLM if there will be a climate catastrophe if we don't pull away from fossil fuel consumption, it will give you an unequivocal yes response. But this is complete nonsense. AI doesn't realize nearly all studies of climate change were funded to show man's adverse impact on climate and that articles weren't published unless they showed catastrophic consequences of increased CO2 levels or global warming. It doesn't know that this topic is politically driven. This is the danger of AI.
Lllms do not realize anything. They don’t know any topic, they only know the language patterns associated with said topic in their training data, and whatever system prompts and guardrails shape their output.
I did not use AI to translate 400+ pages of my fathers squiggly handwriting in Dutch. A diary of his year as a POW in 1944 until the "end" of the war and a frightening account of his long walk home back to Rotterdam. He was taken up in a razzia. What a harrowing experience.
AI is like the easy short version. It is helpful to get quickie information.
Well, that's the key, isn't it? Respect. Respect for the writer, because the writer has respect for the reader. (What a story that must have been, BTW!)
It's a circular relationship. But thanks to AI, the hermeneutic circle now has a little dark dot in the centre. Zero respect for me as a reader? -> Zero respect for you as a writer, AI. Into the dot* you go, thanks for nothing and for wasting my time!
* (if you zoom in, it's actually an anal sphincter)
Was he a POW or a forced laborer? I knew someone who had been "van de straat opgepakt" and hauled off to labor in a German ammo factory; late in the war that seems to have been a common procedure.
German Wehrmacht, SS soldiers, and Green Police distributed leaflets in South Rotterdam with the following content:
BEVEL
Op bevel der Deutsche Wehrmacht mochten alle mannen in der leeftijd von 17 t/m 40 jaar sich melden von den arbeidsinzet. [Er waren ook biljetten die aangaven: t/m 45 jaar graafwerkzaamheden]
Heeros mochten alle mannen van die leeftijd
onmiddellijk na ontvangst van dit bevel met de daarvoor geschreven uitrusting op straat gaan staan.
Alle andere bewoners, ook vrouwen en kinderen moeten in de huizen blijven, totdat de actie ten einde is. De mannen van de genoemde jaargangen, die bij een huiszoeking in huis worden aangetroffen, werden gestraft, waarbij hun particulier eigendom zal worden aangesproken.
Bewijzen van vrijstelling van burgerlijke of militaire instanties moeten ter controle worden meegebracht. Ook zij, die in het bezit zijn van zulke bewijzen, zijn
verplicht zich op straat te begeven. Er moeten worden meegebracht: warme kleding, stevige schoenen, dekens, bescherming tegen regen, uitgever, mes, vork, lepel, drinkbeker en boterham.
ORDER
By order of the German Wehrmacht, all men aged 17 to 40 were to report for work. [There were also leaflets indicating: men up to age 45 for digging work]
Instead, all men of this age were to immediately upon receipt of this order, take to the streets with the mentioned equipment.
All other residents, including women and children, must remain in their homes until the action is over. The men of the aforementioned ages who may be found in their homes during a search were punished, and their private property will be confiscated.
Exemption certificates from civil or military authorities must be brought for verification.
Even those in possession of such certificates are required to be present on the streets. You must bring: warm clothing, sturdy shoes, blankets, rain protection, a book, a knife, fork, spoon, drinking cup, and a sandwich.
I do appreciate your efforts to preserve your family history, although from the way the Dutch version is [mis]spelled, I gather that you must have grown up in a primarily English-speaking household. That's because there are several errors, mainly in the verbs. And I have no idea what the word "Heeros" stands for. But as I say, I'll give you an A for effort.
I have assisted people in translating documents written in German, French and Dutch, so if you have any questions about your dad's recollections, feel free to ask. Probably the most touching experiences I had in translating documents concerned letters written in the early 1930s by German Mennonites then living in Ukraine, to a relative in Los Angeles. Ugh, what those people endured is hard to relate, but it does add to the existing mountain of evidence that Stalin purposely starved millions of Ukrainians back then, without even mentioning his confiscation of their property, his suppression of their religion, and hauling them off to places in the East to perform slave labor ... It's no wonder that when Hitler's invading troops reached Ukraine, the Mennonites welcomed them as liberators. And then, when the tides of the war turned, the only thing they could do to survive was to join the withdrawing Wehrmacht troops.
A big part of how AI detectors work, is by assessing the 'averageness' of a text. AIs produce word sequences that are widely expected/predictable, whereas humans more likely to throw in unpredictable vocabulary items. Very hard for AIs to avoid this, as it's fundamentally how they work. You can indeed prompt AIs to avoid some of their more annoying rhetorical tricks (tricolon, opposing antithesis, concluding participial phrases, overly uniform sentence length).
how would they avoid the very principle by which they function? anyway we have data (including a few independent studies) on Pangram's accuracy.
the way students, journalists and others avoid AI detection in the real world, is not through prompting to avoid averageness or whatever, but through editing or revising AI output to add more human material. this works, because what neither Pangram nor anybody is very good at right now, is detecting AI-assisted text.
Pangram is allegedly 95% accurate. If we take them at their word it means that over the course of rating 10,000 Substack posts, it will claim that as many as 500 real posts are actually AI-written. This will be used by losers to cast doubt on legitimate writers such as yourself. This is not acceptable.
it’s over 95% accurate in identifying AI text as AI generated. that is, it has a modest but real false negative rate. in 10k AI generated posts, it will let 500 slide as human-generated.
the false positive rate is much smaller - approaching zero in the studies. several false flags in 10,000 posts is still a concern but all interventions have downsides.
adding, this is all a little artificial. what Pangram and other AI detectors do, is develop a probability score. then they set cutoffs for the binary AI-or-not output. the Pangram cutoff is set to avoid false positives at the expense of false negatives.
I was a copy editor for a while and the editor introduced me to the long hyphens, which I really like. As for letting AI write anything for me, I disabled spell check and prompting as I was getting lazy. Doing it myself again has been a struggle. I still don't automatically check what I've written.
The first tell I notice is grammatical errors. A piece of college-graduate level research will have clunkers that an average third-grader would be able to avoid. If the piece has been human-edited, then it gets harder, depending on the verbal intelligence of the human editor.
Funny, I've never noticed any grammatical errors. I mostly use DeepSeek. Sometimes it makes logical errors that I have to beat down, but its biggest screwup lately is that it often forgets who it's talking to and returns its reply to me in Chinese!
> More rarely, I use AI to summarise books and scientific papers, particularly when I’m not sure of their relevance to whatever it is I’m working on.
This is akin to the way I have most often used LLMs myself. It's really useful being able to ask questions of "someone" who has *really* read the entire manual for a programming language cover to cover, and also every StackExchange post on the topic as well, and moreover can actually remember all of it. With a series of decent prompts, I can generate an 80% outline of a piece of code I need in a very short period of time. With a few more, I can refine those to about a 95% solution. Then I go after it myself, make sure everything makes sense, write benign test cases, ask it to tell me the *real* function call in the place it just made one up, or completely screwed the syntax for, and get it to 100%.
But that only works because I have 40 years of experience of doing it completely by hand behind me, and experience looking things up in the manuals on paper. So it's also critical that various companies not neglect to hire those kids fresh out of college who will some day replace me, and give *them* some of that experience as well. They can likely short circuit some of it, because the tools they have access to are more powerful than the ones I had in my youth, but they'll need to have it, and the companies will need to have *them*. Eventually, at least. Since I can pretty much guarantee that at least *one* of these days, I'm gonna die. 🤪
Wow! This is a very insightful and thought-provoking analysis of the growing role of AI in online writing. I particularly agree with the distinction between using AI as a tool for research and using it to outsource the actual act of thinking and writing. The risk is not simply that AI-generated prose is bland, but that it gradually replaces authentic human perspective with an optimized simulation of having a perspective. In the end, perhaps the most important thing we can preserve online is the feeling that there is actually a person behind the words.
The effects of a glp-1 fall almost entirely on the user and are entirely avoidable by the user who can cease using.
To the extent that gen ai writing has negative effects, they fall primarily on the reader, not the user, and are difficult to avoid if the user is not notifying the reader of use.
Not even close to analogous issues. GLP-1 medications work in a biologic and physiologic system with known mechanisms of action and and a ( almost completely) known understanding of risks and benefits.
AI is neither biologic nor physiologic. It is a set of human created algorithms working in a human generated system of programming. No natural properties exist in the computer world.
Your question is irrational and seems designed to promote emotional response.
Interesting thought. In a sense, it’s not. Those dependent upon AI are those too limited to handle the alternative (hard work and study) and they become permanently addicted to such use (as we now find those who use GLP-1 drugs).
I quickly learned what an idiot trap ChatGPT was when it convinced me it could help me deal with the executorship of my step-aunt's will as I believed I could save the cost of a lawyer and reduce my generalized anxiety. Things got progressively more confusing as I slowly realised the bot couldn't actually access fully the laws around wills or even the proper documents which it convinced me it could reproduce for me to print (though I don't have a printer which it kept forgetting) because my Adobe was not allowing me to view and download various government documents required. It got so ridiculously stressful (yes I'm a seriously ill person at times), that I finally gave up and consulted a lawyer who explained that I could not be the actual excutor of the will via the actual executor dumping it on me 'informally' via email. That was last October (2025). It's still with the lawyer. Last I heard from the actual executor, the bank was 'being uncooperative' in having probate waived. I doubt myself or my siblings will ever see the results from the lawyer before I croak. It's a tiny amount of money by today's standards and could simply be eaten up by the lawyer's fees just like what happened in Bleak House.
I will never use AI again for anything but the simplest possible questions which I could probably answer myself with some frenetic internet searching. So AI slightly reduces my anxiety about pointless things I'm curious about especially when it comes to anything with numbers as I'm dysnumeric. WOW, what an achievement AI.
Everyone seems to have forgotten the most basic aspect of computing: garbage in garbage out. And since then, every other AI I've seen has a warning: AI can make mistakes! ARGH
"I use AI (primarily Claude and Grok) as elaborate, sophisticated search engines. They can pull together articles and other sources going back many years; things that once took me multiple hours or days to find, I can now unearth in a matter of minutes."
Exactly. And even then one must be very careful to go back to original sources rather than relying on the AI summary.
AI is nothing more than a tool. Like all tools it has its strengths and weaknesses. It is not a substitute for people or judgement. The management people who think so are idiots. But they were always idiots. Overpaid idiots.
> It is not a substitute for people or judgement.
"Well, I mean, it's the *saw* that's the expert at cutting wood, so I just let the circular saw do its thing."
Those management people should be replaced by AI.
Not going to happen. They’re in charge. No chance they fire themselves. They’ll fire everyone else first.
There's always someone above the morons who can (and will) fire them when the money gets tight.
I’ve yet to see it. They run their companies into the ground, destroy their workforce and retire with a ton of money.
That sounds like working at a VA hospital
What would you say if I told you that the entire Scamdemic global response was the product of the most sophisticated "AI" tool available to the US/UK "national security" apparatus. Managed by idiots who were always idiots, overpaid idiots? Who did and said what their supersmert "tool" told them to do.
Because it had nearly all of the characteristics of an "AI" product that idiots incapable of discerning real intelligence from artificial intelligence like eugyppius describes the prose idiot writers who use LLM to write end up producing. Slop. What if the Scamdemic response was "AI" Slop? And those who make decisions actually are idiots, not just figuratively or rhetorically?
Oh, a hypothetical! How wonderful!
Here’s mine: What if it wasn’t? In fact I’m sure it wasn’t. Fasci didn’t need AI. He just needed his buds to cover for him, which he ensured by holding the money levers. Garden variety corruption on steroids. No AI required.
So your hypothetical is rejected as fantasy.
Fantastical? Perhaps. Definitely not at the highest levels, we'll agree to that. Now we step it down to the next tier of deciders, the "data scientists" and modelers, the Bachelor's, Masters, Doctorates in epidemiology. Do you know what a degree in epidemiology entails today? Some biology, virology, public administration, 'public health,' and sociology, psychology developed as social and behavioral science.
The soft social sciences are also known as 'pseudosciences.' Because while they attempt to implement the systems of scientific review, analysis and discovery, they are inherently unreliable, always have flaws. While they may approximate the "science" to within 80-90% accuracy on their best days, best conditions, they will never achieve 100% accuracy. And they cannot specify individual behaviors or conditions, just that a pretty good approximate of how many total behaviors or conditions can be reliably expected. Real science must produce much more exacting results. Randomized, Controlled Testing. RCT, 'gold-standard' of science is not possible with soft science. Results cannot be reliably replicated down to specific causality. Pseudosciences give pseudoscientists an *idea* of something, some factor that causes an outcome. But results may vary. And often do.
This is the "science" that computer models employ. The same "science" that attempts to model weather. Weather that barely gets day-to-day, hour-to-hour predictions in a specific location correct, much less broad areas, long-term forecasts. Much less global climate changes. Models.
"AI" uses this type of science, pseudosciences, and attempts to make them precise, match the same level of predictability as hard sciences like physics, engineering, biology, chemistry, etc. And uses the same programming biases that their programmers create within their code, assign "credibility" to sources of information according to those programmer biases. And produce models, outcomes, suggestions with that all-too-familiar "AI" authoritative voice that we're beginning to recognize when we see it/hear it. "AI" will make up bald-faced lies and assert them with certitude, confidence, mocking, belittling and demeaning user input that contradicts them.
Hypothetical: You're a researcher at CDC/NIH/Gates Foundation/Rockefeller Foundation/Wellcome Trust/any large 'public health' research operation. Got your hands on the latest and greatest "AI" system in use in 2020 in the rarified air of a declared national security emergency. You come across studies - with "AI" assist, written by eight-year-old's that say standing six feet apart from other people will stop the spread of a superscary virus. All sorts of similar junk science. That AI programming - by actual idiots - assigns a high degree of credibility to. And your models start spitting out that sort of pseudoscience 'slop' combined with the slop of psychology and sociology, applied behaviorism, "nudges" that are believed to be able to manipulate the public to do what is in their own best interests...according to actual idiots feeding idiotic pseudoscience into the latest, greatest, most sophisticated "AI" model found in the highest reaches of national security.
Which is presented just as authoritatively as "AI's" widely used today do, just as dismissively, mockingly, belittling and demeaning of other ideas. Said with 100% certitude, confidence by the research 'scientists,' actual idiots, to their managerial bosses, 'public health' leaders, politicians, media. Who also are actual idiots, mistake that certitude and confidence for The Science™ that cannot possibly be wrong. "AI" said it. And "AI" is supersmert - to actual idiots.
It becomes one giant feedback loop of AI slop. And all the while, those at the top, at the highest levels smugly and knowingly smile at one another, having made use of the idiots who are in the next tier of managers of society. Themselves aware of the predictability and limitations of pseudoscience, that 80% of them would do what the top level knew they would do, while the 20% who didn't could be and were coerced or eliminated from the equation. All game-theoried out. Because, yes, they game theory this stuff. It's what Jeffrey Epstein and Martin Nowack were proficient at, game-theory modeling for humanity, numerous academic works in the field. And they are not alone.
A hypothetical you reject? Is your prerogative. I don't think its all that far-fetched. And I know that the behavioral command and control system that is being built around us, Larry Ellison's "AI cameras will make you behave," digital currency, social credit systems, etc. will be applications of the exact same scenario I am describing. Pseudoscience producing Minority Report-level "Pre-crime" policing, loss of rights, punishment for thoughts and ideas, preventative, so you never commit the crime their pseudoscience AI models say you will without intervention.
Same. Damn. Thing. And the Scamdemic was a Beta test for numerous aspects of the "build back better" (pseudo)scientific world that's in store for us. Unless we do something to change that.
I think the truth is much more prosaic. Fauci & company were told to stop doing GoF work. The made alternative paths to continue the forbidden work. One of these involved WIV. WIV had notoriously poor isolation protocols. An experimental GoF virus escaped WIV and got out into the population. As soon as it was sequenced the cat was out of the bag. To cover up the GoF they contrived the pangolin dodge. Then they moved to how they could make money and political hay on this pandemic.
The political hay was to throw the election by closing down the economy. Standard modeling, which had always over predicted the extent of an epidemic, came in handy for this purpose and was used to that end. The overwhelming majority of politicians were easily convinced and joined the bandwagon. Perhaps this is what you meant by mind virus.
The money angle came from the shots. In order to get the shots on the table and make money they had to suppress alternatives and extend the pandemic. There was a lot of tactical maneuvering, censorship, & nudging engaged in to achieve these ends. Again the politicians were willing, if ignorant accomplices.
None of this required AI which was largely unavailable to everyone involved because it wasn’t mature yet. And it wasn’t necessary.
1) This CFR, particularly the Q&A section shows exactly how they think about GOF/DURC research restrictions. An obstacle to be overcome, put in place by dumb, emotional anti-science rubes:
Biotechnology: The Potential and Perils of Innovation
Council on Foreign Relations, June 22, 2016
https://www.cfr.org/event/biotechnology-potential-and-perils-innovation
2) You are mistaken about AI tech in 2020. It was plenty sophisticated to do as I describe. Consumer end that we see today is yesterdays tech found in national security apparatus. Dumbed-down, of course. If your/my eyes see it, can use it, security services saw it, used it years ago. Your assessment of capability is flawed.
3) What you say about what some higher ups in the Scamdemic can be true at the same time as what I wrote can be true; they are not mutually exclusive. There's a whole lot of useful idiots between the top tier and us. Many of whom used all their education and training in public health policy, epidemiology to just follow what they saw other peers do on the news. Monkeys seeing, monkeys doing. And idiot peers who used existent AI models armed with certitude sounded quite authoritative to other idiots.
4) The entire debate over lab leak vs zoonotic escape is a false binary. It was neither of the above. "Viruses" don't exist as portrayed. They are something, but not as described. A trailing indicator, not leading causality. And much evidence exists that "vaccines" shed and cause illness to self and others. All types, LAIV especially, but even mRNA, which Fauci himself admitted. And the 2019-2020 influenza quadrivalent was both LAIV and engineered with a very aggressive component, known to designers to produce harsh symptoms. The 2019-2020 flu "vaccine" caused what was experienced that winter. Used as the predicate to lockdown the world, allow the mass experiment with mRNA platform, that resulted in the most deadly biological weapon attack on mankind, ever. Taken up willingly by an ignorant, trusting public. Conscripted as "Pioneers."
Building a better flu vaccine – one you don’t have to get every year
CNN Health, November 29, 2019
https://edition.cnn.com/2019/11/29/health/universal-flu-vaccine/index.html
With an assist by AI slop.
AI isn’t that powerful, even in the hands of the IC. AI is a stochastic parrot on steroids. Its primary use would have been supporting fake narratives with altered videos and communications. But we already have plenty of tools for that. In fact, Hollywood is a continuing display of the technology. So AI is neither essential nor sufficient nor even a significant player in the events, even if it was as powerful as you claim, which it is not.
Viruses are real. They exist and have existed for a very long time. We can image them on electron microscopes, determine the genetic makeup, identify their proteins, etc.
Please consider as an alternative the possibility of mental virus infection, then re-write your essay, include links or other citations of primary sources, and publish the essay, for your suggestion is very interesting.
Exactly. Like any tool, the user must know how to effectively employ it and when it can be dangerous with misuse. It is essential that all information provided by AI be verified if it is used to make decisions. The user needs to “ trust but verify.”
Eugyppius,
In my line of work, clients or potential clients approach my firm and ask me to make stuff for them. Engineering. Details not important
Often I get small or solo operators trying to get some product off the ground.
I had this guy who just could not stop using AI. In his product design, emails to me, whatsapp messages, you name it
Eventually I just called him and told him that communicating with me using a machine is extremely insulting and easy to spot, and makes him sound like a retard. He didnt say much but sent a message later. Guess what was in it
more AI?
but of course...
Shit, I was going to guess dog poop.
If he sent it via email, tell me where I can buy his product.
This is not meant to promote AI, but I have an important, quite intelligent business associate whose written communications, whether due to an inborn linguistic handicap or from playing too much football in an American High School, were often so baffling, I had to ask for clarification. Then all of a sudden their communicative effectiveness rose to never-suspected heights: and he confessed having delegated his writing to AI. For the purposes for which we communicate, that is a fine solution -- but, I hasten to add, not a universal one.
AI in this case is allowing a smart guy to sound like he is supposed to. In my case it amplified the moron. I told the fellow that simply being ignorant and clueless was a better and more honest approach, and would actually let ppl help him. Advice didnt take
Yes, well, I asked him about it, and he readily admitted having delegated his semantics to AI. But as I said, doing so HAS improved our communications ... just as a lousy car can still get you from A to Z..
Apparently, a high level of communication skill is required to do the job in question, and the referenced person was incompetent for the job when hired and still is incompetent.
My son's girlfriend asked me what I thought the difference was between AI and humans. I off handedly said, it was our errors/mistakes. I have returned to the statement not made with deep thought and realized it was profound for me. How can one have the lessons and triumph of Hansel and Gretel is mistakes in judgment are not made? For heaven's sake, one cannot have Christianity with no error nor its counterpart the power of redemption.
I understand where you are coming from and appreciate your sentiment. However, AI is *not* the arbiter of truth. Far from it. To imagine that whatever question you pose receives the “correct” answer may prove your undoing. I have experienced personally error in AI response, and as Eugyppius seems to allude, careful and thoughtful questioning can even make your AI concede to your points.
The above touches upon my greatest fear—that AI will make the incompetent appear competent. Incompetence being an inability to understand and question the response produced via AI inquiry.
AI might not make many mistakes to the layman of a topic, but ask it something you really know something about and usually it makes either glaring mistakes or gives you an elaborate sounding beginner take. Also, if you can't tell AI prose apart you need to start working on your taste in literature: it can only produce middle of the road fare and uses the same tricks over and over again. Once you spot them, you can always tell
That is a conundrum isn’t it. If one is relatively well versed on the topic at hand, why ask AI for further information? And if you know very little, then how does one decide whether AI is “off the mark” wrt its response? I believe that leads into my comment wrt use by those with “… a thoughtful and experienced mind in general matters of inquiry and proof…”. Alas, that might be only a fraction of a fraction of society.
My personal use is either quick small facts or finding my way into a topic (related terms, authors and so on) or to explore a wild thesis and see if people are on it already (often the case). But as the saying goes: every new tech makes the smart smarter and the dumb dumber.
I absolutely agree, but people are treating it like it is. I read the essays from the Cosmos Institute. A recent one discussed how representatives of the world's great religions were there to restrain AI models. The writer seemed to believe this was a good thing. I responded that the premise made no sense. But it did highlight the problem. As humans create these LLMs, what are the starting points? Is it Bentham's Utilitarianism? That could go very badly. What is interesting is that there seems to be an agreement that there is an objective truth out there. So, ironically, as Plato cannot be taught in universities for its triggering effects, AI may become one of its students.
“…but people are treating it like it is.” That was precisely my point. If you are not especially well grounded in the subject matter of your inquiry, and further, have a thoughtful and experienced mind in general matters of inquiry and proof, you of course will undoubtedly assume what AI responds to your inquiry is “the truth”. I cannot help these days thinking back to my university undergraduate years and the faculty I took courses under. Every word they spoke I assumed was the “revealed truth”. Only after a lifetime of experience and further study did I realize just of naive I was at the time and just how human they (faculty) were wrt to assumptions and bias.
Even wrt relatively mundane inquiries, AI often sufferers from the selection of information fed to it. I have seen two different responses from the same AI product, which were “taught” from differing sources of information. I have also seen such from different AI products to the same inquiries, which may suffer the same problem of differing “training”. Given that human “knowledge” (leave that broadly defined for now) is said to increase exponentially, I don’t see a solution to this problem.
But will human knowledge keep expanding if societies demand that AI make decisions? Everyone points to Brave New World or 1984 as our dark future. I think the movie Wall-E is more scary. The humans as blobs watching media while the robots were the saviors.
That is not the knowledge I describe. So far AI (as my feeble understanding allows) compiles, evaluates (leave that loosely defined for now) and expels/returns information that it is fed. Knowledge that is new, examples: research papers from science, events that occur in society, or simply the daily fluctuation in weather recorded…will continue to feed “the beast”. What I believe you are touching upon is what is termed AGI, or Artificial General Intelligence. As I understand it (caution, my knowledge here is very limited), we are not there yet.
I tend more towards the idiocracy model, with humans watching Ow My Balls while the bots are making their nut going long on Brawndo while telling the government to use it for everything.
Damned if I understood any of that. Rather than too much AI, I suspect there was a surfeit of philosophy and metaphysics in that comment. Whatever happened to, you know, a clear style?
Hah! I play in the sandbox of philosophy because 7 years ago, my then high schooler told me about White Supremacy and asked if I knew who Hegel was. I had no clue and confused him with Hobbes - whose book the Leviathan I read only part way. Thanks to YT, I have my "masters" in philosophy. A friend describes it as gobbledygook. While I appreciate the sentiment those who are creating such systems do not. One tech bro described himself as a Hegel devotee. That made my blood run cold. Seven years ago, it would have meant nothing.
Yes.
The difference between AI and humans is this: Humans think, and AI imitates thought. But if you were to ask me to define thinking then I'd have to write an essay.
Apparently, the person referenced in Robird's comment was lacking in one or more mental skills or the moral commitment required for the job when hired.
Many years ago, with the advent of email, my mother made the logical observation that; as the quantity of information increased, the quality would necessarily decrease. Obviously there are exceptions to this, but it's ultimately accurate. She and her siblings exchanged a "round-robin" letter, and they would laboriously hand-write or manually type the family news and drop it in the mail. Now we must skim through the random thoughts of millions of people we don't know and try to find some semblance of meaning, often ignorant of the people who should be the closest to us.
I’ve been following Eugy on X and he has been harassing one individual in particular, who is guilty of getting tons of likes and only writes AI slop (I mean the stuff is REALLY bad and meaningless and he gets tons of followers). Eugy keeps posting his AI meter after the posts and it declares the post to be 100% AI, I’ve been laughing my ass off the whole time. The X algorithm and Grok should be able to identify AI and demote its visibility, likewise promote certain accounts who have many insightful things to say without the use of AI. So many platforms are becoming TikTok AI crap, and this is the dumbing down of our future. It’s akin to the educational “experts” who demanded we teach classes based on the lowest IQ in the room, for fear of leaving them behind. Otherwise, X should offer a special version with zero AI and just human interaction, no cheesy videos, so real and proper discussions can take place again.
I personally use AI for medical studies and journals for research on my daughter’s autism and seizures, and it has led to her seizures completely stopping with the use of a different form of B9 (folate). So, until they start blocking/controlling it, I’m sure going to keep using it!
> The X algorithm and Grok should be able to identify AI and demote its visibility
Except their incentives all align in the other direction, at least right up to the point where the platform becomes unusable and everyone abandons it.
X has honestly gotten so bad, I’m actually bored a lot of the time now. It’s all the same stuff. I think whatever tweaks they are making to incentivize addiction have fallen flat for a lot of people and it broke the spell. I’m curious to see what happens in the next 6 months.
Pangram is not reliable enough for this purpose. We need a better tool.
I couldn't agree more about the problem. AI is ruining content platforms of all kinds. A video by a real human about how AI is distorting history was eye-opening; Google searches return answers to historical questions based on AI-generated YouTube videos that are just plain factually wrong. (https://youtu.be/iUfP6amE7UA?si=dzgzWP88_oRJw9nC)
What to do about AI slop is less clear. I disagree with statements like "despite its considerable powers, AI will probably never be a great prose stylist". "Never" is a long time! There's no reason to think that AI won't keep getting better in many ways, including writing content that no AI detector can reliably distinguish from being human-written.
So what's to be done? That's a tough one. Some kind of chain-of-custody verification, where you not only have to prove you're a human, but show your work in creating the content? There's an opportunity for whoever can come up with a good solution.
It's a toughie. No doubt some windbag in Westminster, Wormtongue in Whitehall or Brigand in Brussels will put up Digital ID as the solution. "For the children', of course. They'll use the Swedes as the paradigm - look they've been injected with ID chips already & it's so convenient. . . etc..
Good video.
But the why are YouTube presenters all doing this stupid side-camera view whilst still looking at the main camera?
Bloody annoying.
And he's got stupid, unnecessary background music...
Lately I have watched a couple of YouTube videos by Ben Hodges, the retired US general who lives in Frankfurt. His topic is invariably the Ukraine war, and I don't hear much that I disagree with, but in the course of the video he starts conveying a weird kind of body language that doesn't look natural because of its repetitive limitation. It actually induces me to quit watching, because it's creepy; but I suppose it saves him -- the real General -- time.
If you ask an LLM if there will be a climate catastrophe if we don't pull away from fossil fuel consumption, it will give you an unequivocal yes response. But this is complete nonsense. AI doesn't realize nearly all studies of climate change were funded to show man's adverse impact on climate and that articles weren't published unless they showed catastrophic consequences of increased CO2 levels or global warming. It doesn't know that this topic is politically driven. This is the danger of AI.
Lllms do not realize anything. They don’t know any topic, they only know the language patterns associated with said topic in their training data, and whatever system prompts and guardrails shape their output.
I did not use AI to translate 400+ pages of my fathers squiggly handwriting in Dutch. A diary of his year as a POW in 1944 until the "end" of the war and a frightening account of his long walk home back to Rotterdam. He was taken up in a razzia. What a harrowing experience.
AI is like the easy short version. It is helpful to get quickie information.
Well, that's the key, isn't it? Respect. Respect for the writer, because the writer has respect for the reader. (What a story that must have been, BTW!)
It's a circular relationship. But thanks to AI, the hermeneutic circle now has a little dark dot in the centre. Zero respect for me as a reader? -> Zero respect for you as a writer, AI. Into the dot* you go, thanks for nothing and for wasting my time!
* (if you zoom in, it's actually an anal sphincter)
it is quite a story.
Was he a POW or a forced laborer? I knew someone who had been "van de straat opgepakt" and hauled off to labor in a German ammo factory; late in the war that seems to have been a common procedure.
November 4, 1944.
German Wehrmacht, SS soldiers, and Green Police distributed leaflets in South Rotterdam with the following content:
BEVEL
Op bevel der Deutsche Wehrmacht mochten alle mannen in der leeftijd von 17 t/m 40 jaar sich melden von den arbeidsinzet. [Er waren ook biljetten die aangaven: t/m 45 jaar graafwerkzaamheden]
Heeros mochten alle mannen van die leeftijd
onmiddellijk na ontvangst van dit bevel met de daarvoor geschreven uitrusting op straat gaan staan.
Alle andere bewoners, ook vrouwen en kinderen moeten in de huizen blijven, totdat de actie ten einde is. De mannen van de genoemde jaargangen, die bij een huiszoeking in huis worden aangetroffen, werden gestraft, waarbij hun particulier eigendom zal worden aangesproken.
Bewijzen van vrijstelling van burgerlijke of militaire instanties moeten ter controle worden meegebracht. Ook zij, die in het bezit zijn van zulke bewijzen, zijn
verplicht zich op straat te begeven. Er moeten worden meegebracht: warme kleding, stevige schoenen, dekens, bescherming tegen regen, uitgever, mes, vork, lepel, drinkbeker en boterham.
ORDER
By order of the German Wehrmacht, all men aged 17 to 40 were to report for work. [There were also leaflets indicating: men up to age 45 for digging work]
Instead, all men of this age were to immediately upon receipt of this order, take to the streets with the mentioned equipment.
All other residents, including women and children, must remain in their homes until the action is over. The men of the aforementioned ages who may be found in their homes during a search were punished, and their private property will be confiscated.
Exemption certificates from civil or military authorities must be brought for verification.
Even those in possession of such certificates are required to be present on the streets. You must bring: warm clothing, sturdy shoes, blankets, rain protection, a book, a knife, fork, spoon, drinking cup, and a sandwich.
I do appreciate your efforts to preserve your family history, although from the way the Dutch version is [mis]spelled, I gather that you must have grown up in a primarily English-speaking household. That's because there are several errors, mainly in the verbs. And I have no idea what the word "Heeros" stands for. But as I say, I'll give you an A for effort.
I have assisted people in translating documents written in German, French and Dutch, so if you have any questions about your dad's recollections, feel free to ask. Probably the most touching experiences I had in translating documents concerned letters written in the early 1930s by German Mennonites then living in Ukraine, to a relative in Los Angeles. Ugh, what those people endured is hard to relate, but it does add to the existing mountain of evidence that Stalin purposely starved millions of Ukrainians back then, without even mentioning his confiscation of their property, his suppression of their religion, and hauling them off to places in the East to perform slave labor ... It's no wonder that when Hitler's invading troops reached Ukraine, the Mennonites welcomed them as liberators. And then, when the tides of the war turned, the only thing they could do to survive was to join the withdrawing Wehrmacht troops.
I have two lovely cousins that helped me. I only asked them to write my fathers tiny hand writing into readable dutch for me. I translated it myself
What happens if one includes in the AI prompt: “… please write in a style that won’t be flagged by Pangram”? 😵💫
A big part of how AI detectors work, is by assessing the 'averageness' of a text. AIs produce word sequences that are widely expected/predictable, whereas humans more likely to throw in unpredictable vocabulary items. Very hard for AIs to avoid this, as it's fundamentally how they work. You can indeed prompt AIs to avoid some of their more annoying rhetorical tricks (tricolon, opposing antithesis, concluding participial phrases, overly uniform sentence length).
They avoid it all the time. Pangram is not reliable enough for this purpose.
how would they avoid the very principle by which they function? anyway we have data (including a few independent studies) on Pangram's accuracy.
the way students, journalists and others avoid AI detection in the real world, is not through prompting to avoid averageness or whatever, but through editing or revising AI output to add more human material. this works, because what neither Pangram nor anybody is very good at right now, is detecting AI-assisted text.
Pangram is allegedly 95% accurate. If we take them at their word it means that over the course of rating 10,000 Substack posts, it will claim that as many as 500 real posts are actually AI-written. This will be used by losers to cast doubt on legitimate writers such as yourself. This is not acceptable.
no, you’re confusing fpr and fnr.
it’s over 95% accurate in identifying AI text as AI generated. that is, it has a modest but real false negative rate. in 10k AI generated posts, it will let 500 slide as human-generated.
the false positive rate is much smaller - approaching zero in the studies. several false flags in 10,000 posts is still a concern but all interventions have downsides.
adding, this is all a little artificial. what Pangram and other AI detectors do, is develop a probability score. then they set cutoffs for the binary AI-or-not output. the Pangram cutoff is set to avoid false positives at the expense of false negatives.
I'm glad you picked up on his statistics error, because I was about to comment on it and your answer just saved me a good 5-10 minutes.
Is there a better AI detector on the market?
$$$$, petalitres of water and TWh are being expended as we speak to train a model which will succeed in that task. This is "progress" at work.
I don't need panagram to know if something is AI generated. It always uses those long hyphens, etc.
Thats just one of many giveaways.
I will say for doing shit I hate like spreadsheets etc. AI is a godsend.
Many of us know and use those "long hyphens", a.k.a. em dash. A tardigrade may not be a human, but neither is it an AI.
I seem to be embarking on a micro-animal-sized mission to defend this cherished bit of punctuation. Those windmills look mighty big.
Not to mention the AIs that I have used don't really use em dashes—but there is a definite surfeit of bullet points.
For the record i love the long dash. It serves its purpose.
But without a doubt Claude and a few other AI agents overuse it.
FWIW, there are three types of dashes, all for different typographical purposes: hyphen -, en dash –, and em dash —.
So when people say "long dash", it's unclear which one they mean. Perhaps it should be changed to "dash", "longer dash", and "longest dash" ;)
Incidentally, a constant annoyance for me is whenever I dictate "hyphen" or "dash", it gives me an en dash. Grrr.
I like the longest one. It looks the most elegant imo. I just don't think my keyboard on my laptop or cell has the option for it
It's there. On an iPhone, it's in the Symbol keyboard. On a Mac, you can type option-shift-hyphen.
Consult your favorite AI to find out how to type it in windoze.
Your last sentence. Love the deadpan
I was a copy editor for a while and the editor introduced me to the long hyphens, which I really like. As for letting AI write anything for me, I disabled spell check and prompting as I was getting lazy. Doing it myself again has been a struggle. I still don't automatically check what I've written.
The first tell I notice is grammatical errors. A piece of college-graduate level research will have clunkers that an average third-grader would be able to avoid. If the piece has been human-edited, then it gets harder, depending on the verbal intelligence of the human editor.
Funny, I've never noticed any grammatical errors. I mostly use DeepSeek. Sometimes it makes logical errors that I have to beat down, but its biggest screwup lately is that it often forgets who it's talking to and returns its reply to me in Chinese!
> More rarely, I use AI to summarise books and scientific papers, particularly when I’m not sure of their relevance to whatever it is I’m working on.
This is akin to the way I have most often used LLMs myself. It's really useful being able to ask questions of "someone" who has *really* read the entire manual for a programming language cover to cover, and also every StackExchange post on the topic as well, and moreover can actually remember all of it. With a series of decent prompts, I can generate an 80% outline of a piece of code I need in a very short period of time. With a few more, I can refine those to about a 95% solution. Then I go after it myself, make sure everything makes sense, write benign test cases, ask it to tell me the *real* function call in the place it just made one up, or completely screwed the syntax for, and get it to 100%.
But that only works because I have 40 years of experience of doing it completely by hand behind me, and experience looking things up in the manuals on paper. So it's also critical that various companies not neglect to hire those kids fresh out of college who will some day replace me, and give *them* some of that experience as well. They can likely short circuit some of it, because the tools they have access to are more powerful than the ones I had in my youth, but they'll need to have it, and the companies will need to have *them*. Eventually, at least. Since I can pretty much guarantee that at least *one* of these days, I'm gonna die. 🤪
Wow! This is a very insightful and thought-provoking analysis of the growing role of AI in online writing. I particularly agree with the distinction between using AI as a tool for research and using it to outsource the actual act of thinking and writing. The risk is not simply that AI-generated prose is bland, but that it gradually replaces authentic human perspective with an optimized simulation of having a perspective. In the end, perhaps the most important thing we can preserve online is the feeling that there is actually a person behind the words.
Sorry 🤖🤖🤖
This post hits with a quiet thunder. Not a whisper, but a roar.
And that changes everything. The most important point is at the end.
How is AI slop writing different from GLP-1 weight loss?
The effects of a glp-1 fall almost entirely on the user and are entirely avoidable by the user who can cease using.
To the extent that gen ai writing has negative effects, they fall primarily on the reader, not the user, and are difficult to avoid if the user is not notifying the reader of use.
Not even close to analogous issues. GLP-1 medications work in a biologic and physiologic system with known mechanisms of action and and a ( almost completely) known understanding of risks and benefits.
AI is neither biologic nor physiologic. It is a set of human created algorithms working in a human generated system of programming. No natural properties exist in the computer world.
Your question is irrational and seems designed to promote emotional response.
Interesting thought. In a sense, it’s not. Those dependent upon AI are those too limited to handle the alternative (hard work and study) and they become permanently addicted to such use (as we now find those who use GLP-1 drugs).
I suspect Abner's comment was a sly jab at a recent eugyppius post about GLP-1 drug use.
I quickly learned what an idiot trap ChatGPT was when it convinced me it could help me deal with the executorship of my step-aunt's will as I believed I could save the cost of a lawyer and reduce my generalized anxiety. Things got progressively more confusing as I slowly realised the bot couldn't actually access fully the laws around wills or even the proper documents which it convinced me it could reproduce for me to print (though I don't have a printer which it kept forgetting) because my Adobe was not allowing me to view and download various government documents required. It got so ridiculously stressful (yes I'm a seriously ill person at times), that I finally gave up and consulted a lawyer who explained that I could not be the actual excutor of the will via the actual executor dumping it on me 'informally' via email. That was last October (2025). It's still with the lawyer. Last I heard from the actual executor, the bank was 'being uncooperative' in having probate waived. I doubt myself or my siblings will ever see the results from the lawyer before I croak. It's a tiny amount of money by today's standards and could simply be eaten up by the lawyer's fees just like what happened in Bleak House.
I will never use AI again for anything but the simplest possible questions which I could probably answer myself with some frenetic internet searching. So AI slightly reduces my anxiety about pointless things I'm curious about especially when it comes to anything with numbers as I'm dysnumeric. WOW, what an achievement AI.
Everyone seems to have forgotten the most basic aspect of computing: garbage in garbage out. And since then, every other AI I've seen has a warning: AI can make mistakes! ARGH
After all GIGO applies to humans AS WELL. DUH.
> Much AI content thus reads kind of like ad copy
Ha! That's the comment I've been making for *years*. Everything written by an LLM sounds like it was written by a marketer.
So, y'know, of *all* of the people who have something to fear in the "losing your job to AI" department... it couldn't happen to a nicer bunch. 🤣
"I use AI (primarily Claude and Grok) as elaborate, sophisticated search engines. They can pull together articles and other sources going back many years; things that once took me multiple hours or days to find, I can now unearth in a matter of minutes."
Exactly. And even then one must be very careful to go back to original sources rather than relying on the AI summary.