Computers – The PolyBlog

AI testing: The Bad…Time loops, tech support quirks, and drift

April 18 2026

By now, most people have seen some form of AI crop up in their tools. The most obvious one is Google’s search engine, which provides results from its AI mode first in the list. You can go pretty far with that prompt, even asking for image creation, although that’s a terrible place to create images (full imaging tools aren’t really available in AI search engine mode).

In my case, I’ve used it for some research here and there, often against a framework I had in mind. More recently, I’ve had it helping me “test” some frameworks. I design a framework for something I’m building or writing, I outline it and paste the outline into AI, and ask it to challenge the framework from the perspective of say gender equity, under-represented groups, or literacy levels. Something more than a grammar check, something less than a full AI partner. When it’s done, I decide if I want to change anything in my approach.

But I’ve discovered some recurring oddities. Not necessarily bugs, just aspects of LLM-based tools that attempt to translate what I’ve said into something concrete.

Time Loops

About three months ago, I was testing Google’s tools to create an image. I eventually moved to ChatGPT to do the same. And both tools had the same problem.

I input a bunch of prompts. Created some sample images. Iterated a few things. All good. Then I told it to “tweak the image” in a certain way, and it said, “Okay, here you go.” But it was the same as the previous image. There was no “change” or iteration.

Okay, I thought, random glitch. Please regenerate the image with the following changes. Enter, whirr, ding. Same image. Huh?

I would then tell the AI that it gave me the same image again. Apologies, whirling indicator, bam! New image, same as the old. No matter what I did, it would not give me anything else.

It felt like a giant glitch. Or Groundhog Day. No matter what I did, same result. I couldn’t get out of the loop.

At the time, I had NO idea what was happening. Was it me? Was it the AI? Was it my browser?

I now realize it’s essentially a memory issue. Each chat in certain tools has an amount of “context” memory built into it. Once that’s full, loops start happening. Things bog down. In some tools, it will say, “Hey, I need to compact, okay?” and it will crunch your chat and go, “all ready!”. Except you have no control over what it ditched. Images perhaps? Instructions you definitely needed it to remember? Gone. In other tools, it compacts without even telling you.

The AI experts advise that where you had it generate a lot of “assets” (pictures, documents, etc.), it’s better to start your next phase with a clean prompt. You can cheat, though … if you ask an AI tool for a “handover” note, it will generate one you can prompt into the next chat, while it quietly fades into an ignored chat window. Waiting to see if you ever come back.

Google AI mode and ChatGPT seem terrible for this. I hit a lot of loop walls quickly. Gemini wasn’t so bad, but I think that was one of the ones that just compacted on its own. I actually prompted it a few times to save just to be safe. Claude, by contrast, doesn’t seem to have ANY of that happening. It hasn’t got stuck in a loop, and I haven’t seen it compressing/compacting/deleting anything yet.

PolyWogg 0, AI -25.

Technical support

One use case people recommend for AI is technical support. I’ve had four experiences using AI as technical support, and it has done a couple of things okay-to-well, and bombed on others.

The first bomb was on support in a program called mIRC. The IRC part of that is for Internet Relay Chat, and mIRC has been my go-to tool for online chatting since the late ’90s, when I used to be really into it. I have a couple of specific uses for it now, and I installed a couple of plugins recently to automate some stuff. Great, except they didn’t work QUITE the way I wanted, and the default display was in 9-point font. So, I asked ChatGPT how to tweak the mIRC settings for what I wanted.

One of the first things I told it was that I was using version 7.8.3. It has changed interfaces over the years, as well as command structures, so old commands won’t work; just like the voicemail messages say, “Please listen closely to the following options as our menu items have changed.” Okay, ChatGPT said, in its oh-so-confident way, that setting the display font to 16 points was super easy. It gave me a simple command, I entered it, and Bam! Error message. mIRC had no idea what that command was.

I told ChatGPT, it said, “Oh, right, sorry, yes, it’s done THIS way.” Another command, same error. “Oops, let’s do it through the menus, guaranteed to work. Click on DCC / Options / Display / Fonts”. Except there is no DISPLAY option under options. The menus have changed. Took me a while to find where fonts were. Made the change. No help really from chat, I just found the setting. Great.

Except no change. It would change the font for the chat window, but not the popup windows that I needed to tweak. Back to ChatGPT. Reminding it that I was in 7.8.3. Oops, it told me, the instructions were for version 4.3 or something archaic. What? Why? I specifically told you NOT to show me guesses, and to ONLY show me solutions that were validated for 7.8.3. It politely informed me that it hadn’t guessed; it had “INFERRED”.

And thus began my long descent into a deep rabbit hole with AI along for the ride, digging small tunnels ahead of me.

I knew the change could be done, that it wasn’t rocket science, and that I wouldn’t figure it out on my own. I knew just enough to know that either the default font or the plugin font was set too low. No other way for it to be wrong. I knew, therefore, Dr. Watson, that I could either fix the original setting, find a way to override the setting automatically, find a way to change it manually after the fact, or ignore it completely. As time wore on, that last option grew increasingly attractive.

To be fair, mIRC isn’t exactly a commercial application like Microsoft Word. It doesn’t have millions of users. And a user plugin within mIRC? That has even less information about it.

Yet each time I asked a question, the AI tool would say, “Oh, I know how to do that!” Except it never did. It couldn’t find where the default font was set, although I later figured out that it wouldn’t matter, it was the plugin font that was the problem. And it couldn’t figure out how to change fonts AT ALL. Nor could I. I opened EVERY file that came with the plugin. Lots of stuff for settings in the pop-up window, but nowhere where it had a font setting. It seems to be hardcoded in the plugin, alas.

I was undaunted. I knew that if I couldn’t do the first two options, I could at least set it after it loaded. Because I could go into the menu, choose Options / Preferences / Fonts / Font choice. Or something equivalent. It took about 5 clicks to get to where I wanted to change the font. But then if another window opened, I had to edit that one too — another 5 clicks.

None of the options AI suggested worked. Auto-load commands, mIRC scripts — none of them worked — and mostly ended up with the AI tool telling me, “Oh, it would have worked if you were using an older version.” WHICH I TOLD IT NOT TO DO! Grumble, grumble.

I found a workaround — I forced the font menu onto the taskbar manually and then told it to stay there forever; now when the pop-up shows up in 9-point font, I can click the taskbar, the menu opens, I change the font to 16 to 20 points, and it’s done. Super easy, two clicks.

PolyWogg 1, AI -25.

Drifting back to shore

This is a newer version of the loop problem. At least, it seems like it is the same sort of error.

I was trying to get Claude to do an image for me. I wanted to create a badge, with an embroidered edge. All of the AI tools take different approaches to images; some work in specific types of image scenarios, others in different scenarios, and others? Well, some don’t work at all.

Claude NAILED the first part of the badge problem. It gave me a perfect ring on the first try, which none of the other tools did (it uses SVG vectors to handle the geometry, hence why it was so accurate). But when it tried to do the embroidery, it failed completely. Nothing it did looked like embroidery.

I scrapped that idea, moved on. About 40 minutes later, out of nowhere, its attempts at embroidery showed up again in the margins. I was like, “Huh? Did I paste an old prompt?”. So I asked it why it included embroidery in that version. It told me because I asked for it earlier, and the algorithm forgot that I said no to it, so it went back and did it again. It had “drifted” back to the earlier setup. A little weird, so I had it add a prompt component that said very clearly, NO EMBROIDERY ELEMENTS. About 20 minutes later, working much further down in the model, the embroidery attempt came back. I checked the prompt; it clearly said no embroidery. So I asked again, “How?”.

This was a second type of drift. It had analyzed the prompt. And because I had asked for embroidery before (positive inclusion) and now was excluding it (negative inclusion), the fact that I had mentioned it at all was interpreted as positive inclusion. It ignored the “NO” part. I suddenly felt like I was working at Foreign Affairs back in the old days of TELEXes where you couldn’t afford for a word to be missed so you would type NO/NO to make sure one of the “NOs” made it through. I didn’t try that with Claude, because it was now a VERY long chat, Claude was getting on in digital minutes/years, and showing signs of confusion. I reset and started with a new chat, no mention of embroidery. It never showed up again.

I couldn’t find a way around it, other than using new chats. Not sure that’s a win.

PolyWogg 0, AI -2.

That’s the bad news. I was going to write about the tips it gave me for GIMP, but that’s a mixed bag, not all bad. And what really excites me is all the good things it’s done for me. That’s the next post. 🙂

2026: C is for Computers

January 4 2026

There are several areas I want to explore this year in the realm of computers, IT, and websites.

1. I want to tidy up my backups. I’m a bit behind, easy to do, but I need to catch up.

2. At some point, I have to figure out my carousel on the PolyWogg site. I created a pseudo-carousel, large header, and well, it doesn’t work for me. Yet.

3. I guess I also want to do something with the look and feel of the PolyWogg site as well when it comes to blog entries that are “sub-areas” — the main blog pages show perfectly, the sub-areas not so much. I’m missing something small to tweak, haven’t wrapped my head around it.

4. I want to do something with old computers. I have some I want to convert into game players, and I’m going to set up one PC to run 3D printing. It’s not urgent, but it will take a dedicated two or three days to sort it out enough to begin work. This includes re-purposing and maybe even jailbreaking an iPad. I’ve looked it for fun recently, just cuz I wanted to see how hard it is/would be, and you know what? Not that hard.

5. But if I set up the computer to 3D print, I also need to set up the 3D printer. That is a bit more complicated. The new one should be relatively easy, but the old one is a mess of pain.

6 Much more elaborate, I have plans for two apps. One is astronomy-related; the other trivia-ish. But I think they’re going to have to wait for retirement. The effort required to get into programming is just too much overall to dabble.

7. I mentioned in my book list that I want to read stuff about AI, and perhaps it should be here too, although I think it is more for my L is for Learning side of things.

8. I’m also not very proficient in a couple of pieces of software that I need to work with in the future — GIMP will be my basic editor of choice for photos, and I have a full training program to help me get better at using it. At least for the basics, astrophotography will require more work. But I also have ideas for the future that will involve video, and my wife and son got me a green screen setup for Christmas. So I’m mostly ready to go, I just need to double down on what I want to create and figure out which software packages I’m going to use. I have a bunch, but not sure which ones are best for what I’m going to be doing. That may take until late fall to figure out though.

And that’s it. Pretty linear, pretty straightforward.

The duality of digital me

October 20 2025

So, I have two main websites:

www.ThePolyBlog.ca — aka this site, which has a bunch of blogging stuff that I do. There are lots of subjects, and it has generally reflected the tagline / slogan — My view from the lilypads.
www.PolyWogg.ca — aka my writing site, The Writing Life of a Tadpole, which has been primarily been about HR and a bit of other stuff.

I’ve played with the sites over the years, moved stuff around, even debated the locations of certain types of files. However, that’s not surprising…how can the PolyWogg site be about my “writing”, yet I have over a million words on the ThePolyBlog site? Isn’t that writing too?

The funny part is that I asked for advice from friends some time ago, and one who has a marketing background saw it very clearly — PolyWogg was my professional writing stuff, ThePolyBlog was my personal stuff. I didn’t quite see it that way, too far in the weeds, but it was compelling if a little bit too early to commit.

The challenge was what to do with about 200 blog posts that rode the line between the two.

My dual life confuses even me

Take, for example, some blogs I have written about life in government. It’s not exactly about my HR guide, but it’s not unrelated either. I wrote about Phoenix audits, and audits in general, and how they work in government. But then the question — is that me writing about professional topics, as I do intend to do some writing on a series of government-related topics? Or is it me blogging about a current issue that I just happen to know more about than the average bear?

Or some articles about performance measurement or libraries, both of which I will write about in the future. Those are a little more related to future PolyWogg guides, and if I already had those guides written, these blog posts would clearly be housed in the same area as the guides.

Astronomy presents a different challenge. I write about MY experiences with astronomy, clearly a personal area, and thus clearly it should be on my personal site, right? Except I also intend to write a multi-stage PolyWogg Guide that will draw on a lot of those blogs, and like with performance measurement, if the astro guides were already done, then astro blogs would clearly go with them, right? I’d be blogging in support of my astro guide. It gets more complicated when I think about astro PHOTOS. I generally publish ALL of my personal photos on Flickr with links to my website, EXCEPT if I’m doing an Astro Guide, shouldn’t my astro photos be with it? Particularly as all of the photos in the guide will come from the same galleries.

And then just to really confuse things, I do book reviews. Almost EVERY writer out there who writes their own stuff AND also does book reviews keeps the book reviews on the same site as their own writing. Except I don’t really see it that way. My book reviews are just me writing about what I thought of the book, it isn’t something I’m doing to market myself or anything else. So, since they are more like my own book diary than anything to do with writing, I feel like they should clearly be on my personal site. Along with TV, Podcast and Movie Reviews. Seems logical to me.

Except I’m also doing music reviews of decades, which I will turn into books/PolyWogg Guides. So theoretically the music blogs should be with the future Guides and thus on my writing site where the books will be…yet then all my reviews aren’t together. I feel the same way about recipes — they’re mainly personal, but perhaps, someday, I’ll publish a collection of them…is that enough to put them all on the writing site?

I know what you’re thinking

You’re likely thinking that I’m being overly anal-retentive and over-thinking it all. But it does bear some thought.

When I came back from Bouchercon, I came back with a renewed sense of “writing” purpose. I am closer to retirement, and I have very concrete plans. Some of that starts with my PolyWogg site as my primary professional writing site, as my marketing major friend suggested as a way forward.

And here’s the weird part. Over the last year, I moved almost 140 posts from PolyWogg to ThePolyBlog so that PolyWogg was very clearly focused on my HR guide and a little bit on my Astronomy Guide. I spent a lot of time cleaning up the posts, keeping it lean. Then, after attending BoucherCon2025, I realized that a lot of content that I moved away are actually better placed back on PolyWogg and that I should start arranging my menus better. I have more content already written than I thought, across a broader spectrum of topics that will mushroom after I retire.

And if I’m frank with myself, likely before I retire. Or at least sooner than my original planned retirement date, regardless of when my date actually turns out to be. It’s in a state of flux at the moment.

So, this past week, I found 185 posts that I needed to move back. It could be another 300 if I moved my book reviews, which I’m not going to do. Another 300 if I included all my TV posts and reviews over the years, but again, those will stay with my personal site. I could probably conjure up another 30 things that COULD go to the writing site, but I feel like I’m moving towards the writing site only being those things that I intend to publish in some form other than my website, or at least are linked to similar publishable content.

But the truly funny part? I moved it around not that long ago, it made perfect sense to me, and this weekend, with a new “vision”, I moved it all back plus some more.

Even I don’t understand the nature of my digital duality. But I’m starting to, I think. Heck, I even ordered new business cards that say “writer, blogger” that I quite like even if I’m not using them yet.

Yet, partly as part of my retirement plans and partly as a result of Bouchercon convincing me that I need to be more entrepreneurial and perhaps more media-savvy with my approach, I’m working on ideas for a third site focused exclusively on a more marketable niche. I’ve been toying with some separation of content to prevent overlap between the non-fiction, fiction, and media-friendly stuff, even to the point of considering alternate noms de plume and trade names to protect some of my life from the net. But I know too that some of that is a fear that one of the three will fail i.e., I know the non-fiction is solid, but maybe the fiction won’t take off as I hope, or that the media-friendly stuff will fail spectacularly. Old-style concerns in publishing used to recommend using different names for each bit so that any failure in one area wouldn’t hurt your brand in another, but well…I am my brand. I’m PolyWogg. Have tail, will hop and type. Read, ribbit, repeat. If I want to really overthink things, there are MULTIPLE rabbitholes in the three areas to avoid or embrace.

But my first two sites are (not quite) locked and loaded, and I’m working on the third. I may not know ALL the versions of me yet, but I know the core one.

So I was hacked, but I don’t know how

September 25 2025

Earlier this week, I got an email saying one of my social media accounts had an unusual login, but it was nearby, and sometimes that happens normally when my one tool uses a different server, etc. or a bot runs from another setup. Not necessarily “me” accessing, but things that I authorized to access showing up on a server in another city nearby. It usually doesn’t do anything else, and often it isn’t even successful. I have a few accounts that didn’t quite have my latest passwords on them, but they were decent enough.

Tuesday, I went to a coffee shop in Nepean, and my VPN on my laptop wasn’t working. However, I also played an online game on Tuesday night that has a lot of stuff going on to run it. One or the other could have compromised my access, I suppose.

I’ve spent a lot of time tonight rebooting accounts, changing passwords, logging any other access out, and generally being paranoid AF. Of 8 things I use regularly, it compromised the four easiest. The second tier wasn’t hacked, and the third tier doesn’t look like it was even viewed. I still upgraded a ton of passwords tonight using a combination of tools and double-checking 2FA. Most are fine, and are the reason I was able to recover what I did. One remains outstanding, not that critical.

Still, quite annoying and tiring. My protections around the castle engaged when the Trojan Horse was opened, and mostly worked as intended. I’d like to write in more detail, but that would be incredibly stupid.

The bigger question is the vector of attack. One of the vectors may have started before Tuesday, except I can’t think of where or how really. The wifi network would be the obvious idea, or if I had used wifi while I was in New Orleans. But the whole time I was in New Orleans, I only used my phone as a hotspot, no wifi. Hmmm…

The fun that remains is when I go to access an app on my phone or load something on another computer and it says, “NO! You shall not pass!” because I haven’t logged into it with the new passwords.

How was your night?

WTF is going on with Jacob’s PC?

May 30 2025

Jacob has a higher-end gaming PC. Not top of the line, but certainly higher than the mid-range. Great graphics card, decent memory and speed, and a nice curved large monitor.

He comes down to see me yesterday afternoon and says, “Umm…my monitor stopped working.” Huh? Yep, he rebooted, did all the basic stuff, nothing. No signal to the monitor.

At the time, I was working my real job, so no time for much in the way of tech support. I gave him three possibilities:

1. Full shut down, see if the PC has somehow lost its setup info;

2. Try the monitor with a different source;

3. Try a different monitor.

He comes back later to say that the existing monitor works with another laptop, no problem, and the PC itself doesn’t work on other monitors. Excellent, we’ve narrowed it down to the PC. Right???? RIGHT???? Monitor works, PC doesn’t.

Now, there were some confounding variables to add to the mix. He’d been running a new game, and the refresh rate was dead slow. He had tried playing with graphics settings, downloaded a tool from AMD, and after that, nope.

I was initially worried he had fallen for some sort of scam pop-up, but it was indeed all legit. And nothing sounded like it should have screwed up too much, but maybe he lost his graphics drivers. My brain couldn’t decide if the PC would still send a proper video signal if the drivers weren’t on it, but I was wondering if maybe the graphics card went pffft.

I popped over to Canada Computers, where we bought it, and they weren’t busy so I said, “Hey, I might have the easiest fix ever. I think he just blew off the drivers.” Which the guy told me wouldn’t matter. It would still send the basic signal, even if only BIOS info. Huh.

He reached over to his computer for checking things in, unplugged the video feed and plugged it into the PC, added power, and voila, Jacob’s login came up. So, the PC **was** working. Just not with Jacob’s monitor. Or any monitor at the house. Huh?

We chatted about a few other things, but nothing that would give me a lead anywhere. But it was working.

So brought it all back home, plugged in again, nada. No signal to his monitor. We did have a small problem with Windows not being still registered, but apparently unrelated. Huh.

Jacob went off to have a bath, I started noodling. I literally couldn’t think of anything. Then it occurred to me that while we had shut the PC down to “nothing”, we had NOT reset the monitor, and it IS a smarter-than-average monitor. It has some internal memory, auto config stuff, etc. And since it plugs directly into the powerbar, not the PC, it is always “on” at least somewhat.

What if I shut it down too to fully off? I turned off the power bar and let everything go to zero. Nothing on, nothing running, etc., and let it stay off for about 5 minutes.

Then, I turned it all back on, started the PC…and got Jacob’s login on the screen, no problem. After his bath, Jacob reenabled the proper graphics drivers, tested all his normal games, and they all work. The “problem” one still didn’t, but we’ll deal with that on the weekend. The rest is running fine.

I’d love to say I’m a god for figuring out how to reset it, but well, all I really did was turn it off completely before turning it back on. Exactly what we tried multiple times, but as I said, the monitor was staying on and remembering that it didn’t like the previous signal from the PC and thus continuing to block it.

I can’t say I was looking to solve a hardware problem last night. But all’s well that fixes itself.