I’m not alone.
I have a love/hate relationship with AI. Hate might be too strong a word — let’s just say I have concerns. And some of those concerns relate to whether sites like Ask Leo! will survive.
AI is an existential threat. Maybe someday for humanity, I don’t know; what I do know is that the threat is real and current for websites you probably rely on for content… like answers to your technical questions.
We’re losing our reason to exist.
The threat of AI
AI is changing the internet. Instead of sending people to sites like Ask Leo!, search engines and chatbots give instant answers, bypassing site visits and cutting site revenue. Survival? Well, right now it means video, community, and direct support from readers. The future is very uncertain and constantly evolving.
Traffic is down
I talked about this a few months ago. The original business model of Ask Leo! (and many other sites across many different topic areas) was this:
- Write quality content.
- Write it in such a way that search engines pick it up (aka “SEO” or Search Engine Optimization).
- Have the search engines present links to that content when people search for related topics.
- Hope people click through to the site to get their answer.
- Have advertising alongside the content on the site to generate revenue.
- Profit?
AI summaries in search engines and AI chatbots have basically tanked that model in the last 24 months. They no longer present prominent links for people to click on, but instead present summaries and answers. People do not need to click through and visit my site.
Fewer visits mean lower ad revenue. In my case, it was bad enough that I removed all third-party advertising from the site. There was no point in continuing to annoy the folks who made it here with ads that weren’t helping support the site.
Help keep it going by becoming a Patron.
Related
Changes for Ask Leo!
Changes are coming to Ask Leo!, including the removal of outside ads. I’ll explain why this (very) scary step is necessary, how search engines and AI are changing the game, what it means for the future of the site, and how you can help.#182543
Yet website hits are up
Ironically, I recently had to increase the size of the askleo.com web server1 because it was getting pounded with page requests.
Not from real visitors, of course. There are two culprits at play.
Search engine spiders. Even though they’re not presenting links as much, they continue to scan the site for updated and new content. How they use that information varies based on the search engine involved. Sometimes they present links to content relevant to what people are searching for, like the old days. More often, though, they use the content they find on my and related sites to generate the summaries and answers they present.
AI spiders. There are many AIs out there right now, and they’re all crawling the web, slurping up content to train their large language models or provide real-time information when people ask questions of AIs. It’s the same model as search engine summaries, though more extensive and complete. And while there are often references to source material, there’s little incentive for people to click through.
So, yeah, I had to get a bigger (read: more expensive) server to feed my content to the various spiders and bots that are scanning the web.
Why let them?
The most obvious question is: why let them? There are various mechanisms I could use to prevent bots from accessing the content on my site. It’s one reason you’re seeing more “Are you human?” tests as you navigate other sites on the web yourself.
Website owners like me face two scenarios.
- Block the bots and know that AI and search engines will not use my content, will never reference or mention me, and will never send people my way.
- Allow the bots and hope that AI and search engines will use my content in such a way that people get the answers they need, and hopefully, a few of them will click through or check a reference and find my site.
It’s not a great choice.
And that’s why I say the fundamental model of publishing useful information online may be dying. What’s my incentive (other than altruism) to keep publishing? Particularly if I have to spend still more money to keep the server from being overwhelmed by bots?
What’s the incentive for anyone to continue to create useful content online?
Related
How I Use AI at Ask Leo!
I write all my own articles, but I often use AI as a tool along the way. I'll walk through a recent example.#168449
Irony: I use AI myself
I’ve written about this before as well: I use AI as a tool more or less constantly. I don’t have it write content for me (hence the “Written by a real human” at the top of every page), but I use it to generate eye-catching images, prompt me to dig deeper into ideas, and make generating these articles easier.
And I use AI when searching for answers myself. I try to click through to references presented (I mostly use Perplexity, which is great about including references to its source material), but I don’t always. Sometimes the AI-generated answer really is all I need.
I’m not complaining that the current state of affairs is wrong, nor am I saying it’s right.
It is what it is, and it’s having an impact.
And that impact is going to get bigger.
AI-generated content
Here’s a scary concept:
The quantity of AI-generated articles has surpassed the quantity of human-written articles being published on the web.
– More Articles Are Now Created by AI Than Humans – graphite.io (an SEO/AEO company)
There are questions about whether real people see those articles and whether they show up in search engine results or are used by AI. (AI being trained on AI-generated content is another, separate, scary concept.)
The fact is, it’s happening.
And as some have commented, you are currently seeing the worst AI we will have going forward. AI is only going to get better.
Does it matter where the answer comes from?
This is like my customer support position of some years ago: if someone understands my problem, and I can understand their answer, and their answer actually resolves my problem, I don’t care who or where they are.
Understandability and accuracy matter. If those fail, then it’s a bad customer service experience. I think we’ve all experienced that at one time or another.
Pragmatically, is AI different? If it gets you the answer you need, does it matter that it’s AI-generated?
To be sure, there are many issues in the background. With human customer support, it’s about jobs and working conditions and so on. With AI, it’s about where those answers ultimately came from.
But as a user, when you need an answer, are you going to turn down the right answer because it came from someone’s idea of the wrong place?
Again, I don’t know.
Written content today. Tomorrow?
Honestly, what’s saving Ask Leo! today is video. Every written article has a video posted on YouTube where I cover the same topic. If there’s something to demonstrate, I demonstrate; otherwise, I discuss (not “narrate”) the article at hand.
Including this one. By the time you read this, there’ll be a video of me talking about the issue on camera.
AI will probably be able to meet or surpass the quality of much written content in the not-too-distant future. I’ll keep writing, and I’ll keep pointing out that I’m human, but I know it’s a losing battle.
The same will happen with video someday. There’s some amazing AI video being generated right now, but it’s not on a par with videos created by real people discussing and demonstrating topics in the tech space. Yes, AI is being used as a tool — for example, someone whose English skills are not up to par can use AI to transform their voice into something more understandable. But it’s still human-generated content.
If you see me on camera, that’s me.
But I’m not sure how long that’ll last. I’ll say we’re good for a few years while AI catches up to the current state of human how-to and Q&A video creation, but catch up it will.
I’m not sure what happens then.
Related
8 Ways AI Can Help You Every Day
Discover how AI can actually make your life easier. From summarizing long articles to rewriting tricky text, answering vague questions, or even helping you write better emails — these eight real-world examples show how tools like ChatGPT, CoPilot, and Perplexity can save time and boost confidence every day.#179862
The Future
I’m not the only one facing this situation. I’m not even the first. Maybe the topic of technology is a little harder for AI to replicate. We’re certainly seeing AI in other spaces further down its evolutionary path than we see in tech.
Creators are responding in several different ways.
Giving up. I hope not to do this, or at least not for a very long time. But someday, the incentives and revenue we rely on are likely to disappear. I’m not sure when or what this will look like for me.
Exploring alternate presentations. This is what I’m leaning into with YouTube and my online course creation. AI’s not there yet, so it’s very much worth my time and investment.
Building community. This is what I see happening a lot outside the tech space. It’s all about building direct access to a resource you trust. Newsletters, Substacks (and equivalent), patronage, and members-only access are all built on one-to-one relationships, bypassing all the things AI is competing on. It’s why I’ve been a little more vocal about my own patronage options.
Something else? Many creators and entrepreneurs continue to explore different ways to do what they do in a way that competes with AI less or complements it more. I don’t know what this looks like, but again, it’s something I’m keeping an eye on.
It’s not just about me
I’ve used myself as an example throughout this essay because I know my issues, and you know what I do.
I am extremely privileged that while Ask Leo! is absolutely a business with a bottom line, paying me a salary and also paying three assistants, I won’t lose my home or go hungry if it goes away. (Same with my assistants, I believe.)
The same is not true for other online creators. Many are terrified to see their livelihoods threatened by the changing landscape. Some may adapt, some may figure out how to make it work, some may pivot, and others may end up giving up on a dream they’ve invested much of their lives in.
Do this
This is not an anti-AI screed, and I’m most certainly not saying AI is evil or that you shouldn’t use it. It’s a tool, and a very useful one. It’s just that it’s having a dramatic impact on various industries, including my own.
With that in mind, here’s what you can do.
- Click through to references. You should check AI’s veracity anyway, since it’s still hallucinating from time to time.
- Consider supporting individual creators you appreciate. That could be as simple (and cheap) as a free newsletter subscription, or you could go further with paid subscriptions and patronage.
- Share the creators you appreciate with others. Word-of-mouth (or “word of mouse”, as one colleague put it years ago) is very real and very powerful. Particularly in the age of AI, recommendations from real people like you carry more weight than they ever did.
And even though it’d be easy to read everything above as gloom and doom, I remain excited by what the future holds. Truly.
Change is always challenging; it’s about finding how best to participate in what is to come.
HI Leo,
I use AI to generate letters or change medical jargon into English. But you answer questions that are not asked, at least by me, because I don’t know enough to ask. Can AI read my mind like you sometimes do? I’m 81 and have been using PCs for 45 years, so I know a little of how things work and then you come along with an article that shows me how to be even better. You generate that, not AI. So people who are curious and always looking to be better need guys like you who can speak at their level of understanding and suggest better and faster ways to solve problems. I never even saw that stupid windows key until I read your articles on their use. Now I can’t live without those shortcuts. You showed me…and I never asked, because I didn’t know enough to ask. Don’t sell yourself short.
It’s not a question of Leo selling himself short. It’s a question of AI ripping of content and repackaging it as their own and putting content creators out of business. If you like Ask Leo!, become a Patron. It not only gets you around twice as many articles, it makes it possible for us to continue producing Ask Leo! articles and give individualized answers.
Hi Leo, I’m with you all the way. I tend to ignore these AI generated answers to my search queries (mainly technical) as I don’t trust them, I prefer to go to websites I know. But even these might now be AI generated, though hopefully reviewed before posting. That’s fine for people that understand how it works, but there vast majority of users out there just take the first answer they’re presented with and don’t know how to distinguish between good and bad answers. I think this whole AI thing is adding to the dumbing down that’s been happening for a while. Yes, it can be a great tool of used wisely – but I worry about our capability to think for ourselves.
Well said, I agree!
Good Article and timing. A friend I know is a new college professor and is having problems (the entire school faculty as well) with students abusing AI to do their coursework. As this college is taught online, there isn’t the recourse to have students write material in person, as a High School teacher I know does – Hand written essays….
It does seem that critical thought, and even basic math skills are out the window these days. I’ve had multiple instances where when paying cash and the bill may be and odd amount, say six dollars 25 cents, I’ll turn over a twenty and one dollar and a quarter. And have the cashier politely inform me that I’ve overpaid and that the twenty will cover everything. They don’t seem to grasp that a 10 and a 5 would provide exact change…
Hrrumph. These kids these days! It’s been an interesting ride. I’ll be curious what the internet looks like in five years.
When I taught college courses, the students would turn in plagiarized work and a quick Google search would bring me to the same articles the students plagiarized. Now we need AI tools to discover AI plagiarized papers. It’s a game of whack-a-mole and we’re way behind.
I’m a retired community college instructor, but before I retired, I well understood the problem AI posed via plagiarism. That’s why I hated doing instruction online (during COVID) because it was impossible to monitor students as they wrote their assignments. And I did not trust AI used to detect AI-generated assignments! Not to sound like a Luddite, but the larger problem is that students (and the public) are losing critical thinking skills due to AI. AI will soon become an “oracle” everyone will worship instead of thinking for themselves. People are lazy.
I am also just at 80 years old and we had the same issue back when calculators were introduced in the early 1970s. When I was a teenager in retail, I gave out change by using addition, a simple concept no retail clerk now understands. Nowadays, they need calculators to tell them what change is needed and are perplexed when the computers are down.
There was a huge debate whether to allow calculators in school classes for just that reason — can’t do simple arithmetic in your head any more! In the end, everyone now has a computer calculator in their pocket, and there is no stopping AI either. We will adapt to the new and in the process loose some of the old.
I remember when I worked at a drugstore soda fountain (anybody old enough to remeber those?) My boss taught me to start with the cost and count the money up to the amount handed me. It took me 5 minutes to learn. Later when I was teaching middle school and college math classes, I banned calculators for a time in eaxh class to get them to learn how to think. They hated it, but I promised the answer would be a round number so they wouldn’t have a panic attack.
Most people also no longer know how to harness up a horse to a buggy anymore either. The skills required for daily activities inevitably change over time.
Thinking about the 20 USD +, you could say upfront ”if I give you ‘X’ amount you can give me exactly a 10 and also a 5 dollar note, and in that way I won’t have to take a lot of bulky coinage from you”
I don’t blame youngsters for lack of thought whilst handling cash – credit cards, QR codes and automation, are edging the proportion of transactions settled in cash nearer towards zero.
Looking at a list of Linux apps yesterday I noted one for Abacus – still hope… !
Indeed I have also found AI generated search results useful, at times useful enough that I had no reason to check the underlying sites. Today, AI is relying on content that has been (mostly) created by humans. If the current state of affairs discourages humans from creating useful content, then where does that leave us. Certainly AI will be less useful, it will have less to feed on. It will have killed the geese laying the golden eggs.
Good point about killing the goose that lays golden eggs. You only got one thing wrong. It’s not “mostly”. It’s all stolen from creators. Ai may be copying from other AI, but ultimaely, it all comes down to content created by people. AI is a fact of life we’ll have to learn to adapt to.
I’m surprised that your volume has been so impacted by AI. I have always accessed your web page, specifically the recent posts, to see what I can learn. It has been much appreciated for many years. With all due respect, I also often do searches for computer related technical questions but rarely get hits on your web site, so again I am perplexed by the recent loss of volume related to AI summaries. .
We used to get many more hits via Google before they changed their algorithm. Ai has made that worse by giving answers based on the work of many creators instead of linking to the creatpr’s websites. This is a new form of plagiarism. Leo’s and other creators’ answers are still getting out there, but AI bots are taking the credit and the revenue.
Mark,
That’s an interesting comment you made in your last sentence “… and revenue.” Since AIs are owned by someone other than yourself, they are either redistributing the revenue you used to get, or devaluing your product practically to zero thus destroying the revenue. Since sooner or later, AIs will be smart enough to tell the difference between original “human made” content and AI generated content, the prospects of human content creators should eventually rise again.
Dear Leo. Although AI may threaten your hits for those looking for specific answers to tech issues, your COMMENTS, OPINIONS and PERSONAL APPROACH to the different questions and issues are UNVALUABLE and AI will need a lot of time to catch up with the warmth and personal touch you provide, specially for those, like myself, where age and technical non-savyness come together. Please keep going!
The problem is AI is making it harder for people to find Leo’s personal approach answers if it internalizes Leo’s and other creators’ answers and spews them out as its own.
AI seems to fit the definition of plagiarism. It combined the information from several sources and publishes them without citation.
It’s not legally considered plagerism, but if a student of mine came up with the same results not using AI and not citing their sources it would be plagiarism.
I see it less like plagiarism and more like what it would look like if a human read a bunch of articles, and then generated a new article on the same topic. But there are most certainly issues.
Either way, it’s a problematic gray area. I see it more as the bot reading the source material, taking note, and then weaving it together in their own words. That’s how I wrote papers in college. Mine wasn’t plagirism because I cited my sources.
I recently searched for info on some topic, and the first two pages of results were all AI-generated sites, NONE of which had the information I wanted, and with content that was increasingly garbled (from feeding on each other) as I went down the list. I never did find whatever I was looking for.
Bot-generated sites have been an irritation for 20 years now (I remember when the business of creating bulk content switched from human writers to machine-built regurgitation) but it was never like this. And I am already very tired of it.
It’s going to keep proliferating so long as ad impressions make money, and as Youtube demonstrates (with the many bot channels collecting ad revenue) human visitors are not required. So it seems to me the real cure is reducing the market for ad impressions. (Do your part. Use an adblocker.)
Meanwhile, we need a reliable way to flag and separate human-generated from AI/Bot-generated, but given AI can easily call itself “human-generated” just the label is not enough. Maybe some sort of certification consortium that rigorously polices itself??
It’s paradoxical, but the solution to identify AI created content will probably be AI 😉
There will never be a “solution”.
Whether the tools built are AI or not, the AI will improve to avoid detection, the tools will improve to detect better, the AI will improve to avoid detection, the tools will improve to detect better, and so on and so on and so on.
In most cases, inadequate or unhelpful AI outputs are the result of poorly crafted prompts.
“Most”? No, I don’t think so. Many, for sure, and just like getting better at crafting search queries, getting better at crafting prompts for AI is becoming a very valuable skill.
I’d put a lot of the blame on erroneous and false information on the Web. Like the old adage GIGO – ‘Garbage in, Garbage out’.
I haven’t used AI. I have a number of trusted experts that constitute my Go-To friends & web sites. Part and parcel to this is staving off brain-rot: asking questions and being spoon fed an amalgam of information stifles thinking and creativity. Reminds me of the old adage: give a hungry man a fish and he’ll eat for a day; teach him how to fish and he’ll eat for the rest of his life. I’ve been hunting, fishing, and growing for a considerable time even though grocers are less than a day away. I’ll continue to hunt and fish on the web for info I seek, and keep my brain well oiled in doing so. BTW, AskLeo! is one of my trusted experts. Hang in there, we’ve yet to see the silver lining on the AI cloud.
I would like to avoid using AI as far as possible, but how? Maybe Leo can give us advice?
My advice is not to avoid it, but rather use it wisely. It’s a powerful tool.
I see the issue as a lack of ethics with those who design these AI systems. I recently read an article in The Atlantic about a company named Common Crawl and the executive director, Rich Skrenta. Though the aim of the article was about web crawlers going around paywalls and scraping copyrighted materials, one comment by Mr. Skrenta stands out to me: “You shouldn’t put your content on the internet if you didn’t want it to be on the internet.” https://www.theatlantic.com/technology/2025/11/common-crawl-ai-training-data/684567/
That comment tells me that AI companies seem to think that every website connected by the internet is fair game for their purposes and that they feel no concern about what it takes to put it there in terms of effort, time and expense for those who create and maintain the sites. All they seem to be interested in is figuring out a way to make a profit from the work of others and calling it their own. If it causes issues for others, that is their problem.
From social media to AI, developers never seem to slow down to think through what the effects of their ideas will be and install guard rails to reduce any negative effects before making their ideas reality. Then they wonder why there is a backlash to their “wonderful” ideas.
Skrenta said “You shouldn’t put your content on the internet if you didn’t want it to be on the internet.”
What he meant was, “You shouldn’t put your content on the internet if you didn’t want me to steal it.”
Hi Leo, I generally take AI generated info with a pinch of salt, I see it very much as a summary of so many sources it can lead you astray, it reminds me of friend at junior school who started teaching me to play chess, not long after I asked my dad to play a game. He had being playing it casually for years, he then corrected all the aspects my friend had misunderstood. So to me AI generated is still very much a child and is probably re-digesting other AI generated text now, call it ‘knowledge incest’ once started it could generate an even more distorted view of what was factual information, incest can have bad effects in human make up, where will it lead to in AI?
Leo, you wrote:
“And that’s why I say the fundamental model of publishing useful information online may be dying. What’s my incentive (other than altruism) to keep publishing? Particularly if I have to spend still more money to keep the server from being overwhelmed by bots? What’s the incentive for anyone to continue to create useful content online?”
If AI is the problem, Leo, my suggestion is to try to make it a part of the solution. Feed AI your concerns, and ask it if there is any solution it can provide. Consult various different AI’s, and see what suggestions they come up with. Hey, who knows, maybe one of their suggestions will actually be useful!
———
Yesterday, I dared to ask ChatGPT-5 an unthinkable question: Is there was any way to activate Policy Editor on my Windows 11 Home edition computer. Windows Home isn’t supposed to have Policy Editor available, so this would be some trick — and a highly useful one, if only it works!
Well Sir, ChatGPT-5 furnished me with a lengthy *.BAT file for me to try. Sometime this week I intend to boot up Macrium, back up my system, and try that puppy out.
And if it works? I’ll post it here in full!
Yeah, AI really is starting to generate some very useful content…
On of the nuances that I think is important when using AI is accuracy in the question. For example I would expect that what you want is the Group policy editor. (There’s another policy editor as well, I think.) It’s the GPE we miss on Home editions. I’m curious if you get a more accurate answer by asking a more accurate question.
One thing I’ve ruminated on, but not enough to put into lengthier words, is that using AI is forcing me to ask better questions. That’s not a bad thing.
If that hack doesn’t work, Policy Plus is a great replacement for Home versions. GPE has a few more features that support Enterprise installations, but for home and small business Policy Plus is just as good. I use it myself.
Leo, don’t go!
I do want to offer caution about having AI write letters for you.
Writing is thinking. Finding the word that says what you really mean to say is important for thinking clearly and for being authentic.
Authors always say that writing is difficult, and it is, but it is worth the trouble.
Using AI to think and write for you is like having your exercise bike running but not being on it.
Wonderful post, Jane, thanks, especially the last sentence.
When I was a kid, my mother would sit me down to write thank you letters after my birthday. I would have appreciated AI then. 😉
I get your point. My mother did that too and I hated it.
Now the children send me a quick text that says “Thanks, Aunt Jane!” That is enough.
They don’t have to find a card and an envelope and a stamp and go to the post office. That is drudgery.
The text is easy for them and makes me much happier than an artificial letter would.
My main use of Ask Leo is not to get answers. My objective is to get filtered information from a trustworthy expert editor so I can keep up on what are hot topics.
For as long as I’ve used the Internet, and I remember the ARPANet, it (the Internet) has been analogized with the American Old West, a place where law and order don’t exist, so maybe it’s time the world brings law and order to the Internet, similar to how we govern everything else, perhaps starting at the UN level. For something like this to work, on the scale required, the member nations would first have to grant the UN the authority to enact laws to define and govern what we may consider to be international crime in general, and specifically for the Internet. Next, the UN would have to create an Internet police force, charged with investigating Internet crime (as defined under the new UN laws), and tracking down then arresting the perpetrators, so they can be tried in the International court, which in and of itself would have to be expanded, and authorized to judge these cases.
I know this is a lot, and the scope of what I’m suggesting may well make it impossible to enable, but it’s the best solution I can come up with right now. What do you think?
Ernie
So you want the Illuminati to make the rules. 😉
Just kidding.
But seriously, there would still be countries that don’t sign that treaty.
That Wild West meme (in the original sense of the word) is a movie trope. Towns in the West had town governments, law enforcement, courts, and even gun control.
I know, but that was only in the towns which were a relatively small portion of the western landscape. The vast spaces between those communities were relatively lawless, although not quite as markedly so as is the Internet of today. With that said, I was using that ‘trope’ to make a point, not to describe the old west as being completely lawless. Even the Internet isn’t completely lawless, as can be proved by the steady stream of arrests being made by the FBI and other similar law enforcement agencies around the world. The primary point of my original post was that the world needs a globally scoped law enforcement agency with the authority to make arrests wherever the locate these globally scaled criminals, and take them to the World Court, or something similar to be tried for their crime(s). Nations are thinking on a national scale or smaller, while these Internet criminals and those who operate on a global scale aren’t, and in my opinion, that’s a problem.
Ernie
“a globally scoped law enforcement agency with the authority to make arrests wherever the locate these globally scaled criminals, and take them to the World Court” sounds very 1984. Law and order is good but consolidation of power like that would be a disaster. It’s an authoritatian’s wet dream.
Don’t hold your breathe on that one.
If humans no longer create content – what will AI do ? – After all – AI is just “using” existing content – not creating any new content – just variations and condensed versions of existing content. So AI tomorrow will end up being based on what AI wrote today – in other words – self-confirming in a never-ending narrowing process.
A valid concern. It’s been discussed, but no one really has any good answers, at least not yet. I think a lot of people are kind of assuming content will continue to be created, even without today’s incentives. I’m not so sure.
AI trained on AI has a bucket-load of issues.
Another problem is that AI will train on AI and it will be like Chinese telephone or a 10th generation photocopy.
Finally vanishing up its own fundamental orifice.
I very much agree with most of the posts so far and I’m very much an oldster like Ernie. Having only a rather limited income, I make most of my donations at the end of the year after seeing what my finances are. I’m sending off today an increased annual donation to Ask Leo!
Rather than monthly, is there a way I can set up a one-time annual amount and perhaps become some level of patron in further support of your much appreciated work by you and your team?
Don’t forget deep fakes
https://www.facebook.com/share/r/1B9Gv4JrHF/
I disagree with your decision. I think you should provide revenue producing advertising in a side panel. Such panels are not distracting to me as long as there is only one narrow panel and the ads are not flashing and not videos. You are running a business.
The problem is that side-panel ads (I’ve done them in years past) generate SO LITTLE revenue as to not be worth it. The reason that so many web pages are plastered with ads is because that’s what it takes for advertising to work these days.
I generally use Google Chrome to do searches. Many times (not always) I will put “-AI” in the search bar prior to the inquiry. That way, I don’t get the AI responses. I look for specific information that I can easily cite, as needed. AI is handy, for me, in a limited way. I almost turned off the AI feature on my laptop but couldn’t quite do it.
I am in a lonely time-machine, preferring the human touch whether on a laptop, phone, making a business call, or what have you. Your information and approach are highly valued by many of us. Hopefully, you will be able to maintain a strong presence in the tech world.
I find they often hallucinate and come up with something out of left field. Even more often they misinterpret my insructions and give me something I didn’t ask for. Bottom line: if you don’t understand the subject matter very well, the answer will do much more harm than good.
The way I use it mostly is I enter both the question and the answer and ask it to improve the grammar and the clatiry. Often, it mistakes the answer for the question, ot it just gets it plain wrong.
Don’t trust, verify.
Hi Leo, I have been reading your emails and some of the articles they refer to for many years. Thank you for what you do. Although I have known of AI for several years I have only really used it once (to paraphrase my objection to a proposed massive warehouse developement). Until reading this article, I hadn’t realised what it means for content generators like yourself. Thinking it through to a logical future, if all humans give up generating content, the quality of content generated by AI will diminish and eventually become imagined rubbish. I’m glad I still have my education and an enquiring mind.