AI is making teaching history much harder

In our Introduction to Research classes we’re teaching/taught how to distinguish between and value primary, secondary and tertiary research sources and what plagiarism is and is not. Those who fail to abide by these rules” will always thereby expose themselves. A bit like the saying, “You can’t fix stupid, but the red hats sure make it easy to spot.”

Max, just keep doing what You do!!! Those who truly follow you, know of Your work, love and respect it. I think we all may have cringed with the announcement of AI. It scares me.... man kind, imho, was already teetering on a downfall academically (and other areas), but putting AI front and center with no way to hold it accountable, will continue to demoralize education. History already has been challenged, and will continue to be so. People like you who put the truth into your content will always have high value. Thanks for all the recipes along with the history lessons. Carry on!

Unfortunately, this seems to be the case for EVERYTHING on the internet now. I tried searching for some recipes the other day and I got huge piles of sites I'd never heard of, many of which had the same weird errors in their "recipes". Same thing for travel information sites. It's reaching the point where actual research on the internet is becoming prohibitively difficult for me. Fortunately, I'm old enough to remember how to do research with books, but this bodes poorly for a lot of people.

The internet has never been the most reliable research tool. Books still reign supreme.

AI slop is right. I’ll stick to authentic content creators, but I definitely feel for you as you try to find honest research. Chin up, and march forward into history, as ironic as that sounds, Max.

I agree 100% with you Max. And it's not just history. A few science blogs I follow have, unhappily, started depending on AI and their content has quickly downgraded to the kind of fodder that you used to find in supermarket checkout lanes. I 100% believe that there are places where using AI is totally appropriate, automated manufacturing lines for example. But I also 1000% believe that 99% of the places that AI is being used today are totally inappropriate. And it's being forced on us, whether or not we want it. I'm sorry it's affecting you and hate that you are part of the "scrapee's".

Just as mass produced furniture has increased the value of hand made bespoke furniture due to being better quality, so too will the mass gunk of Ai generated slop increase the value of the genuinely researched and compiled article ...someday. Maybe... ...I hope.

I agree completely. I think this problem has been brewing a long time, and AI has simply capitalized on it. The internet was already full of factoids that were spread widely, quickly believed by people with no historical knowledge and no common sense. Their bullshit detectors are completely nonfunctional, hence them believing things that, to you or me or probably most people in the comments here, could quickly identify as false even without knowing anything about the subject. But then AI came along and, like my drunk uncle with high verbal fluency, is capable of producing a near limitless number of coherent sentences expressing entirely incoherent ideas. Rapidly, people looking to make a quick buck have used the former to capitalize on the naivete and ignorance of most people, resulting in the mess we have now. I can't know this for certain, but I suspect that in countries with better educated populaces, AI has not had such a significant impact, because people can better recognize AI-generated content and ignore it. It's discouraging, and while I hate the idea of creating significant regulations on a technology, I have quite literally no other ideas of how to counteract this problem now that we're here. A better educated populace would likely not fall victim to the current trend, but that's a multi-decade solution to an immediate problem. At the risk of sounding like a Luddite, I fear that AI is one of the greatest problems facing us today -- if it's competent and intelligent, it will rapidly lower demand for white collar and artistic work; if it's incompetent and unintelligent, it will rapidly disseminate falsehoods and make it even more difficult than it already is to find the truth. Either case is bad, as I see it.

I'm worried, too. Researching data is a big part of my day job and, frankly, something I do a lot for fun as well. (My favorite way to spend a Friday night is with a few glasses of good wine and a deep rabbit hole!) I am constantly running into AI-generated falsehoods, many of which seem to be referencing other bad AI-generated data. It scares me that Bill Gates is pushing the narrative that AI will replace doctors and teachers in as little as a decade.

Also, you did an amazing YouTube video with my favorite guys. Ocean liner Designs, Michael Brady and I couldn’t believe it when I saw you on my other favorite channel and Ghost Town living with Brent Underwood!! Been following those guys for a while and fell over when did a compilation. Seriously, AI can’t one close..!!!

There’s no one like you out there Max! You are the personality and delicious sense of humor that is unique. Always and forever a fan!!!!!

I believe that the issue with AI on this topic is that it makes it easier to get a simple, digestible summary of "research" that can be generated in less than half a minute. It's the path of least resistance. I remember my grade school teachers calling me out for using wikipedia for that exact same reasoning. With so many things fighting for your attention and time nowadays, I can see why people would turn to the quick and easy answer. It's distressing, but not surprising that people will just take whatever is summarized at face value to make a point and get on with their day (or another video).

This stresses me out too! It's so hard to avoid AI and I really haven't seen any strong argument for generative AI besides its ability to break writer's block. It very much feels like blockchain all over again. so much hype for so little benefit.

To put it simply, these large language models are made by optimising certain math thingy for one particular goal. The goal is to look as similar to already seen human-made texts or paintings as possible. That's it. To LOOK real. Yes, there are layers and layers of maths and all the texts in the world, but in the end this is it. So AI should not be trusted with anything related to any kind of truth. Ever. And it is incredibly sad and frustrating that so much people fail to realize this. But at least all this takes me to trust this chanel more :)

I hear what you're saying, Max. I personally like AI (it's very helpful with brainstorming, for example, and hallucinations can actually be useful in a creative context when you don't need verifiable facts), but I agree that the amount of crap it generates is overwhelming and difficult to counter. I'm a libraran in a school that has chosen to embrace AI (though not everyone is in favour of this decision). Educators have the ability to shape where AI goes and what it is trained on (for example, the rise of a "Fairly Trained" certification). Some AIs are more ethical and accurate than others. My goal as an educator is to help students form the critical capacity to form a best practice of ethical use, understanding that when it's appropriate to use AI will shift according to context and need. But that's a huge battle in a landscape where students don't even necessarily understand the results of a Google search or know how to use a print book's index, never mind AI. The sheer volume of things they need to know to have a functional level of information literacy is overwhelming all on its own.

This happens all the time on FB with Archeaology posts too. It's very discouraging. That old quote about those who don't study history are doomed to repeat it becomes terrifying when the study leads to falsehood and lies. Someone powerful enough to inject their revisionist history has only to commission enough posts to influence the AI towards whatever views they wish perpetuated.

Thanks for being open about this. As a software engineer with 30+ years of experience who has also been following AI development closely for even longer, the way people are treating these LLM chatbots as “oracles” is terrifying, and shows a horrific lack of critical thinking on the part of so many people who should know better. Not only are they just autocomplete on steroids, they don’t even have any mechanism to model things like “correctness,” *all* they have is statistical prediction of next-token likelihood (just like an iPhone keyboard). Anyone promoting them as something more is either stupid, deluded, or trying to sell you a bill of goods.

I teach at an information school. I'm adding this to my syllabi to help explain problems with AI.

They've started pushing ChatGPT at my job. Our legal department hesitates to use it as it is often wrong. A co-worker tells me there are programs used by schools that can detect when AIChat is used. Not allowed. Apparently, there are telltale signs. So, maybe there is hope. People can make corrections under those erroneous Youtube videos. Keep up the good work.

Let me start by saying I'm sorry. As an AI researcher, I too am appalled by slop. The one ray of hope I can offer is that explainability (why does the AI think that is true?) and hallucination elimination are areas of active research in the AI community. We are actively looking for wise models that we can trust.

We feel you. Please know that we appreciate all the work you put into researching food history. Honestly, I preferred AI when it was limited to science fiction.

AI is ruining the internet, and especially YouTube. The Large Language Models are designed to sound natural and convincing in human conversation. They make little or no effort to be accurate, but they are very good at sounding right. I'm afraid the problem is much worse than you imagine and is quickly flooding social media and search engine results with meaningless babble.

Keep the Faith Max. We are counting on you !!

The worst thing about this rant is that it doesn't feel alarmist at all. I'm a graphic designer by trade, and even finding stock images in the past year has become harder. At least Adobe Stock allows you to filter out the AI-gen slop, but the number of images it's filtering out is getting larger and larger (often half or more of the search results) and, because Adobe gives the same value to both a real photo and AI-gen crap, I can only see that getting worse, too. All I can keep hope about is that people still value people-generated stuff, whether it's art or historical facts, over AI-gen uselessness. Ever since Google AI told me the safe temperature to cook chicken to is 38°C I just don't even trust the most basic of stuff it tells me.

It's not alarmist, just accurate. People have been sounding the alarm for a while now. The way AI works in its current form is basically fancy autocomplete, guessing based on what's statistically likely in all of the random internet slop they've scraped without asking. It's similar to that game where you start texting on your phone and use the suggested next word to see what it comes up with... Except... AI is starting to scrape slop from other AI in addition to un-fact-checked general internet nonsense, so now it's *recursively* terrible, and it's getting worse. I'm told that grammar checkers—always worse than spellcheck—are now actively wrong.

I won't support the use of AI for anything but data analysis, where it's useful. Im systematically deleting my socials, I refuse to give a cent of ad revenue for them.

No, my friend, you're not an alarmist and it's not going to get better any time soon. Without getting into the weeds, this is not a random occurrence but a well-planned move to separate humanity from information, particularly history. The value of history is to learn from our past mistakes and not repeat them. Information is power, and Those Behind This absolutely do not want humanity to have either the information or the power to rise to our full potential. Trust your gut (pun intended!) and if it doesn't feel right, don't ingest it!!!

Around 1982, I remember having a discussion with a very smart man who said that computers would eliminate the need for humans to create art and literature. I argued that without the input of humans, originality and truth would vanish. It makes me so sad to see this actually happening. Even sadder is that we so willing accept this garbage as fact.

About a month ago, I read a terrifying article about AI (I wish I'd saved it) that stated the general attitude of AI developers is they have no responsibility for what their programs generate. Accuracy and fact-checking aren't even considerations. At most, their job is to make sure their software works and is easy to use. Everything else is buyer beware. Unfortunately, so many people are so used to being spoon-fed information that they just lap up AI-generated slop without questioning its authenticity. There have always been half-truths and non-facts in circulation (you know you're getting old when you remember the old emails of "fact lists" that included "The only month that didn't have a full moon was February 1865"), but what was once a trickle of nonsense is now a tsunami. People who have never learned how to process information are at the mercy of people who don't care, and that endangers all of us.

I agree with you and am sorry this is taking some of the fun out of what you do. Between deliberate efforts to bend the truth, and AI being unable to tell truth from fiction, and algorithms designed to funnel users into echo chambers, it's no wonder so many people are ill informed and yet so confident in what they "know". It makes the work you do and the integrity with which you do it even more valuable. In addition to the fascinating history you share with your audience, you set a good example for other content creators!

I feel ya Max. My husband will have a convo. with AI without blinking an eyes. I'd rather talk with a real person :) BTW, history is a fave of mine, esp. historical bios. Daddy taught us history at the supper table. One date that will always stand out: 1066.....it's a battle involving a king and a conqueror. Since you have done so much, I bet you already know!

Rant welcome and accepted. Keep doing what you're doing Max. We love it!

Sadly the people who are funding AI don't have the slightest care for history, or for anything else other than the next quarter's profits. Human welfare, human history, and actual human development don't make the line go up, and in fact (le gasp!) often interfere with the line going up, so they're at best collateral damage. Please keep doing what you do, the human race desperately needs people who are passionate about actual facts and true creativity.

I couldn't agree more! As a researcher (of many things history-related, but not food - I leave that to MM) I always go to primary sources whenever possible. When I need to rely on a secondary source, I always find a way to confirm what I find. AI isn't even a secondary source, maybe it qualifies as tertiary or quaternary. Who knows where the information comes from! People are far too quick to accept and believe anything they read on the internet. What is out there is just getting worse with AI.

You are right on the money. This is going to get worse and worse until something tragic happens to start the far too late regulation and control of this technology.

You are right to rant.

AI will speed the dumbing down of the human species.

As a high school history teacher, the past couple of years it's been harder and harder to trust ANYTHING students create outside of the class because of AI. I stopped assigning writing outside of class. Even in class, I have to monitor what they are doing and basically lock down their computers to just the sites I want them to use during class. This is one of MANY reasons I was quite happy to sign my retirement papers this week (going into effect at the end of the school year). To celebrate, tonight we're dining in a French restaurant in Detroit :)

My facebook feed was FLOODED for a while with AI recipe pages all sharing their "recipes" for Strawberry Cake. It's still flooded with AI, but the number of fake recipe pages seems to have diminished, at least for my feed.

You are so right. One of the problems is that people are abusing the use of AI. I use it like an assistant. Check grammar. Sentence, structure and to rewrite something using a different tone. Sometimes I like it sometimes I don’t. I also use to research like a search. But like with a search I don’t so ways trust it. AI has a place, but somehow there needs to be a of controlling it.

I too share your alarm. I am particularly worried as I observe my grandchildren growing up in a world that is, more and more, manipulated by the motives of quick profit and dubious fame. AI is the latest powerful tool of that mass movement. In the field of food, it is not just the increasing spate of false history, but also sites and promoters who are publishing AI generated recipes and hacks which are not only bad, or faulty, but are often dangerous. Ann Reardon of How to Cook That has tested and exposed numerous faulty and dangerous AI generated recipes, yet the content proliferates exponentially. I am grateful for content researchers, like you and Ann, who are fighting the good fight to preserve knowledge.

You are absolutely right. I have heard similar concerns from another creator. We love your work and we appreciate you!

I know without a doubt that your concerns are entirely valid and difficult to ignore. I'm 66 and I recall how utterly awful (and exhilarating) it was when I was in honors English class in high school in Baltimore and we were asked to write a paper on some theme in Shakespeare's plays. The caveat was that one of our sources had to be a primary source. That meant we had to find a library somewhere in the area we could visit that had a book or pamphlet that was dated from the 1500s. I picked witchcraft in Tudor England. I was fortunate to have an amazing resource, the Enoch Pratt Free Library. At the time, they owned a copy of "The Malleus Maleficarum" (The Hammer of the Witches) and I was able to sit in their rare books room and carefully sift through the pages to locate the quotes that I needed to support my thesis. Understand, at the time, there was no online catalogue. There was no online. I had to take the bus to the library and rummage through the extensive card catalogue until I happened upon it. Research was beyond difficult before the advent of the Internet, as you can surmise, and thus why only the most dedicated academics found success. They were the ones with access to the inner sanctums of libraries or archives. They weren't called ivories towers for no reason. Librarians were mages to me. I wanted to be a mage. Now anyone can pretend to be a mage, even someone who only exists in a digital realm. AI can eventually be a great boon, but it is not ready for prime time. And Azuria is absolutely correct: we all must learn critical thinking. I was taught critical thinking throughout primary and secondary school. In my schools, we were asked to analyze current events news articles. I applied what I learned to analyze what advertisements were trying to sell me and with what techniques, including subliminal messaging. Advertisers knew it was important to create fear (especially FOMO), anger or inadequacy in a person and then tell them how their products would solve that problem. Sometimes those problems didn't exist until the ad appeared in the magazine or on the television screen. Yes, it is a complex and challenging. I have just finished the book by John Green, "Everything is Tuberculosis" and in it he writes: “Looking at history through any single lens creates distortions, because history is too complex for any one way of looking to suffice.” AI is a tool for looking at history, but it creates distortions, like humans have done for centuries. I hope you can navigate this digital labyrinth and not be overly dismayed. We are with you!

I really, truly believe that stuff is poisoning our society in multiple ways and I refuse to touch it. Tech companies are waaaay overselling its abilities and shoving it in our faces to keep their investors happy, but at what cost to us all?

I feel ya. Sometimes I ask Google something and they tell me something that I'm certain is incorrect. Or sometimes they give contradictory answers if the same question is worded slightly differently. I hope society quickly adapts to second (or even third) guess their conclusions after considering data... and "answers".

I ran sets of fake laboratory patient results through CHATGPT it and asked 'explain the meaning of the tests and tell what is wrong with this patient'. It would correctly define most of the tests, but it would skip over some and occasionally it would reverse the meaning of a result (ie say that a high HCT meant that you had anemia). It behaved poorly insofar as providing a diagnosis was concerned, sometimes ignoring tests or sets of tests (ie saying that the diagnosis was a 'low Potassium' when the K was given as normal in the result set I provided). I suggest a humorous video on the most extreme examples of 'fake history' that you have found.

As an illustrator who has seen job prospects dwindle SO far to nothing that I've had to change career paths : your concerns are not idle. I am hoping that soon we will in general turn away from what a friend of mine calls the "content blender" that AI is.

I do know how ai text generation works and it's doing far less than not fact checking.. ai tools like that have no reasoning ability at all, what they do is generating text that sounds good. They're literally just mashing text together..

AI is like a mediaeval monk copying a manuscript, then letting the quill do the writing instead. Just keep doing what you are doing, that way you don't need to justify yourself for saying the truth.

I have students try to correct me, the person with the degree, with what that google ai thing says. And it’s just insane some of the things it comes up with for Latin

I could not agree more. I work in software development, actually in applying AI to specific fields, and I have to argue, forcefully sometimes, with our Product team to message our solutions as not magic and not even "trust but verify" but actually "don't trust and verify", it mostly works, but I see misuse/abuse/I-don't-care-if-it's-true-or-not, is everywhere, all the time.

You are a realist, not an alarmist! Too bad the false information is making it harder for you to filter through what's real and what isn't. I appreciate your diligence, Max! The amount of fake information and "how to" videos being posted is alarming and some of them are potentially dangerous to those naive enough to try without thinking it through

there's a you tube chanel with ah istory bent that as all AI and puts out 4-5 videos a day every one of them absoulte shit. It helped me make up my mnid that there is nothing worht while about AI.

The same thing is happening in the mushroom foraging world (which I'm a part of) - all of these AI-generated mushroom guides/books/articles. The content is laughable, and would just be annoying, except for the fact that if people use these as real guides, they might get themselves killed - so it's serious business. I never would have thought that this could get in the way of you making your living with all the work you do; hopefully, people will learn better how to detect AI-generated content and ignore it (and that you can somehow ban all the AI bots on your social media).

I agree with so many comments. You are not being an alarmist, but see what is happening in regards to AI and its use at this time. I appreciate the time and effort that you put in to your research, but I also understand the frustrations when AI can pop in and seem reliable. As a teacher, I try to teach my students that they ha e to double and triple check information and not just rely on one source. Thank you for everything that you do. And don't worry about having to rant on occasion. Hugs to you, Jose, and the kitties.

Well said! Generated slop is ruining the internet, which makes me so sad. I'm a fellow internet creator, and I share a lot of similar concerns in my corner (writing and illustration).

Categorically not alarmist. I block every AI account I'm served, but the drivel is getting worse every week. Too many people wanting a shortcut to ad revenue instead of putting in the work. Despicable. All the worse because it makes things that much harder for the real artists and writers trying to get real and meaningful things out into the world. :(

I hate that this is happening, but unless there is some sort of certification that says the information is backed by real human fact checking and editing, I don't see it ending anytime soon. This is exactly why some science podcasts footnote their transcripts with references. It's a good practice but it's very difficult to duplicate for channels run by individuals.

I'm seeing one of the other flavors of this which is over dependence on AI in live systems. I'm a software developer and I am increasingly seeing peers that are less experienced overly rely on AI produced code that does not work the way they think it does. 'Vibe Coding' which is putting a prompt and running the code that the AI generates even if the code doesn't do what it is supposed to is becoming more common. I had someone turn in code this week that broke production and they cut and pasted what the AI said about what their code does. When I explained that this was incorrect and showed what their code actually did, they said I was wrong. I had strong 1984 flashbacks but... almost a decentralized/democratized version of it.

Max, my family has seen all your episodes. We discuss the history you present and we make use of your recipes and cookbook. We learn from you and we are grateful.

I wish you were being an alarmist, but I think you're correct in being concerned. I love history. Love love love it. I'm 68 (eek) and I studied history by reading book after book. I look back at my education and compare it to what it is now ... I have no words.

AI is such an issue in no small part because it isn't actually capable of understanding anything. "Intelligence" is a misnomer - it's basically a statistical output tool which searches for the likeliest combination of words based on the query, but it isn't capable of discerning what's a "fact" and what's "bollocks" because it isn't even trying to do that. The reason it can't value particular sources over others, determine reliability, or just give you essentially a search result is because it would make it significantly more obvious that it's just copyright theft on a huge scale. Using copyrighted data as a statistical tool isn't necessarily against the law (you can't copyright a fact, as an example), but you *can* copyright the specific wording used. If there was, for example, only one source for a piece of information and an AI tool spat out a copy and pasted section, that's infringement. You also can't just rephrase it slightly and say that isn't plagiarism (which is what catches a lot of students out at universities!) "Generative" AI isn't actually able to generate anything. It's basically just summarising every single result you see on Google, but with no capacity to assess whether something is right or not - because being "right" isn't actually part of the programming. And to me, that's probably the most terrifying part. I'm glad you mentioned the AI slop, and think you're right on the money.

AI ain't any better at science.

Facebook, YouTube, etc. Could stop ai spam accounts if they wanted to. Those idiots are going to poison their own platforms to extract one last dollar in the current quarter. It's really just end stage capitalism.

It’s a thoughtful set of comments. Funny enough I am an AI scientist and I train GPTs. The way you described the issue is actually very astute. We have to think of AIs like ultra smart kids that can go ahead and learn anything presented to them - but they lack the critical thinking and ability to discriminate. What I do in my job is to train ‘specialist’ AIs that are beginning to appreciate that they can’t believe everything and that they have to test themselves for accuracy Sadly, why would would anyone spend the effort to do that with AIs when they can just use ChatGPT out of the box ? Mind you - how many humans would similarly go on YouTube and pretend to be historians or scientist by simply parroting things they saw elsewhere without fact checks Speaking as an AI geek, the human species must absolutely make it our business to provide the chain of custody for every piece of information we consume. And we must all learn critical thinking. Meanwhile - be hopeful. There’s much good out there to be achieved yet.

Thank you Max!! All your hardwork shows. I don't think you're being an alarmist at all. If we're supposed to learn from history, but all the accessible history is incorrect, what a mess we'll be in. Also, imagine how made it is for people researching health issues, etc. Big hugs.

I totally appreciate the rant, AI is a seductive tool but still just a tool that can be used well or poorly. I use AI from time to time in my job to help accelerate research and provide a framework for whatever I'm doing, but see it as just a starting point versus a solution by itself

Well said max. The whole worlds gone mad.

And unfortunately, to go with that, a worrying number of people are taking all of its output at face value. It can only produce a coherent sentence, not a *true* sentence, and you can't really give a text generator the ability to determine what is and is not a fact. That would be actual AI, in that it would have intelligence and be able to interpret things rather than just printing text - and it's most likely decades away. Just like it was decades ago. Genuinely appreciate that you're not on board with it, Max, and I'm not surprised considering how much slop it's got to be putting in the way of actual research.

Thank you Max, I am of the generation when we actually used encyclopedias. And I still prefer books, I don’t trust the internet and don’t get my news on Facebook. However , I know I can trust Tasting History with Max Miller and I am grateful for having you do it.

Oh Max, as a member of the choir I can only exclaim, with a nearly 15th -16th century religious enthusiasm, "PREACH!!!" As a writer for a company (I'll leave out the details) AI hallucinations nearly got us in the kind of hot water that libel suits are fully cooked in (thanks Marketing Dept!). I take comfort in people like you who validate my zealous belief in good old fashioned research and fact checking! Keep up the good fight and keep the faith!

It’s awful and ruinous to cumulative human knowledge. Maybe I don’t need to know who invented a specific kind of screwdriver, but more than half the info out there now that comes up top of list on searches is AI garbage. How to clean some specific surface? Mostly bunk results. The only real crowd source of knowledge on “doing things” anymore is old net forums, which thankfully I’m old enough to think of as an option. And that’s just research. I’m an illustrator, and the sheer scale of theft and devaluation of human labor and creative endeavor is crushing.

Now put this in terms of Politics, and you can see why we are on the brink of civil war.

This makes me all the more grateful for your dedication to the true history of food and its ability to shape civilization. Thank you for all you do and for all you share.

I completely agree with you. AI is becoming a big problem due to its inaccuracies which only reinforce its own inaccuracies. The more an incorrect "fact" is told the more AI thinks its correct and repeats it. There is no good mechanism to "correct" this problem. 😞

It is alarming. While AI does have many potential benefits, it currently has a lot of flaws. It will be up to people who love history and facts to uphold the truth. Does it take time? Yes. But that is the price one pays. Some things cannot be easily Googled and searched online.

this process is known as "enshittification" and while AI has shown some remarkable benefits in some fields, it is truly terrifying in others. Like, how do we ensure that the AI is grabbing from only primary source materials and not made-up bullshit? Do the humanities have to roll out their own AI assistant to ensure researchers are getting proper answers and context? AI isn't going away, so how do we work with it and fight the dumb?

I share your views. You are absolutely correct and not being alarmist at all! You are not the first historian I have heard with these concerns. I have seen multiple YouTube videos from others sharing these same feelings. You absolutely should share your views in your newsletter. Sadly, the majority of people who need to read this, likely won’t. I’ve cautioned my daughter for years about trusting Wikipedia articles, and AI is so much worse!!!!

It's definitely concerning. But, in my opinion, it's a symptom of a poor education system. Barring that being rectified, I think it's really up to people with their heads on straight to speak up and not let the tsunami of bullshit discourage us.

I can see it getting a lot worse and, unfortunately, I don't see a turn-around. I have family members that believe whatever they read on social media. I can filter out the AI noise for myself, but so many around me don't seem to have the same ability. As long as there is a profit to be made from AI, we are not in a good place.

I agree that you're not an alarmist. There really needs to be regulation to slap a label on AI-generated content so that people are more aware of how prevalent it is, and then go do their own research to verify what they're seeing/hearing.

Why not make it an advantage for yourself. Why not putting in a section of debunking AI fables? It'll give you decades of source material. Also, that would be a good bit of archaeology and a way of helping making people aware how important fact checking is.

I am in the medical Publications industry and chatgpt is getting involved in writing manuscripts reviewing manuscripts, and creating pharmaceutical ads, which I find terrifying. The program has "hallucinations" in that it plugs in invented information when it finds a gap. In medical research this is very scary, as references and data can be fictitious, which can affect drug approval. Writers have to be very careful with this.

I would like to assure that I am an actual human. And a slightly drunk human at that. :-)

AI just wasn't ready yet. They're training it on a circular dataset so it's just making itself dumber. The problem I have is that it's becoming harder and harder to tell when something is AI. It's getting better at faking being real. And that's the part that's scary. People are treating it as fact when it obviously isn't. I work in public health and this is becoming a give issue in my field as well. I very much appreciate creators like you who put in the time and the effort to try and be as factual as possible!

Agree that you are NOT being alarmist. The trend is unsustainable, and must collapse eventually.

You are not alarmist, but in fact showing proper amounts of concern and it is sad that so many are not.