Bot or Not?
It’s Getting Harder to Tell What’s Real

Every so often, I check the comments spam folder on my blog before emptying it. Usually, there is no question about why a comment has landed there.

Most of what lands in that folder is exactly what you would expect: generic praise, awkward wording, questionable links, and comments attached to posts that are months or even years old. Recently, I came across one that made me stop and take a much closer look.

It had been left on a post I had published only about two weeks earlier, and whoever, or whatever, wrote it seemed to have actually read the post. Not skimmed the first paragraph. Actually read it.

The comment referenced the 15-day trip, the number of pieces I packed, the outfit combinations, my reverse-order packing strategy, the workout clothes I never wore, and even the difference between what I thought I might need and what I actually used.

In other words, it sounded like the kind of comment a blogger hopes to receive. Except WordPress had put it in the spam folder. And once I started looking a little more closely, things got interesting.

Not the Usual Spam

Typically, spam comments that show up on my blog seem to gravitate toward older posts. I have no idea why a five-year-old outfit post suddenly attracts a fresh round of spam, but apparently it does.

This one behaved differently. As I noted earlier, the post was recent. The comment was specific. The writing was polished. And there was none of the usual “Wonderful information, please visit my website” nonsense.

For a moment, I wondered whether Akismet, my blog’s spam filter, had simply gotten it wrong. It automatically filters out a lot of obvious junk before I ever see it, while the comments that land in my Spam folder are the ones still worth a closer look.

But the more I looked at this particular comment, the less it felt like a typical reader response.

For one thing, it was long. Very long. Rather than reacting to one or two parts of the post, it read more like a synopsis, moving through point after point and neatly summarizing what I had written.

Then there was the information attached to the comment itself.

The name was listed as BioTrendy, the email address was a generic info@ account, and the website linked to a domain that I had never heard of. There were also no previously approved comments from that source.

None of those things, by themselves, proved anything. Together, though, they were enough to send me down a little rabbit hole. So I started digging.

A search for BioTrendy led to a Polish website that describes itself as a marketplace/catalog for health and drugstore products. The site includes supplements and other products related to things like cholesterol, menopause, weight loss, skin care, sleep, joints, and vitamins.

Tangentially related to my niche more or less, but not exactly the audience I would expect to be deeply invested in how many pairs of shoes I packed for a 15-day road trip.

Looking at the website didn’t uncover anything that explained who, or what, had written the comment. It didn’t even tell me whether BioTrendy itself had anything to do with submitting it. Someone could have been using the company’s name and website.

The likely purpose of the comment wasn’t especially mysterious. Spam comments have long been used to generate backlinks. What made this one different was the comment itself. Had a live human actually read my post and written that response, or had a bot or AI tool done it for them?

And that was when “spam” started to look a little different to me. I will likely never know whether that comment was written by a person, generated by AI, created by a bot, or some combination of the three.

The uncertainty is interesting. What I find troubling is that my first instinct is now to suspect a non-human source rather than simply accept the comment as genuine.

Then There Was An Email

Not long after that comment caught my attention, another suspicious message showed up, this time in my email junk folder. It claimed to be from Meta and warned that my advertising account had received multiple violations and was at risk of being permanently disabled. There was, naturally, a large “Submit an Appeal” button waiting for me to click.

This one was easier to question and investigate. The sender was noreply@appsheet.com, not an address connected with Meta. The greeting was the generic “Dear Advertiser,” and although the message referred to multiple previous warnings, it did not identify my account, a particular ad, a date, or a specific violation.

There was also something slightly off about the presentation itself. Facebook, Instagram, Meta, and the supposed advertising violation were all bundled together in a way that looked official at first glance but became less convincing the longer I looked at it.

Next, I did what I have learned to do with all messages like this: I ignored the button and went directly to my Facebook and Instagram accounts instead. Everything there looked normal, so the email went into the trash.

But then I found myself asking essentially the same question I had asked about the blog comment: who, or what, had written it?

Phishing messages have been automated for years, so there is nothing particularly new about sending thousands of them at once. What has changed is how easy it has become to generate polished, convincing language.

A prompt with a few ingredients, such as “Meta account violation, urgent appeal, final warning,” can generate hundreds or thousands of variations that sound more professional than the old typo-filled scams. Was the message written by a person, produced with AI, or pulled from a template?

Through a combination of all three is a possibility as well. And once again, there was really no way for me to know.

This wasn’t the first time recently that an email had made me suspicious. When I received the initial collaboration offer from LINTICO (see LINTICO: Doing Linen My Way), a company I had never heard of, I questioned whether that message was legitimate too. I did some checking before responding, and in that case the offer turned out to be real.

That experience is part of why I don’t want to treat suspicion as proof. Sometimes the message that makes you hesitate is fraudulent. Sometimes it is legitimate. The challenge is figuring out which is which.

And Yet Another Example

I’ve been experiencing a similar kind of uncertainty while reading a few of the blogs that I’ve followed for years. Every so often, a post appears on a subject or topic that seems well outside the blogger’s usual interests or experience, and the writing doesn’t quite sound like the person I’m accustomed to reading.

The posts aren’t labeled as contributed content or as sponsored, but that doesn’t necessarily mean anything deceptive is happening. Bloggers are known to use guest writers, freelance help, contributed content, and AI for all sorts of reasons. But when I know a writer’s usual voice and subject matter, a noticeable departure stands out.

I’m not suggesting there’s anything wrong with writing about a wide range of topics. I certainly do. The difference, at least to me, is the personal perspective. Even when I wander into unfamiliar territory, I’m usually writing about why the subject caught my attention, what I’ve experienced, and/or what I think about it.

It’s that personal element which seems to be missing from the posts that have puzzled me. And that leaves me wondering.  Did the blogger write the post? Was the content supplied by someone else? Or had they simply relied so heavily on AI that their own voice got lost along the way?

And once again, I find myself asking the same question that started this whole post: who, or what, am I actually reading?

So, Bot or Not?

More than likely, I’ll never know who, or what, wrote the blog comment that kicked off my dive into this “Bot or not?” exploration. I absolutely know the Meta email was bogus. As for some of those familiar blogs with unfamiliar voices, I may never know whether the writing came from the blogger, someone else, or a heavy reliance on AI.

Not every question is going to come with a clear answer.

We have more tools than ever to help us screen, verify, research, and question what we see online. I’m glad to have them, and I’m certainly willing to use AI as one of those tools. But none of them can make every judgment for me. A filter can be wrong. AI can be wrong. And sometimes there isn’t a definitive answer to uncover.

So maybe the question isn’t always “Bot or not?” Maybe it’s whether something feels trustworthy enough to click, respond to, believe, or simply keep reading. And increasingly, I think that requires paying attention not just to what something says, but to how it got in front of me in the first place.

What makes you stop and question whether something online is genuine?

Stay well and let’s keep in touch! Subscribe, follow, friend, pin … all options are available. I look forward to hearing from you soon.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Note: This is an unsponsored post; however, some or all of the links to clothing items may be associated with an affiliate program, where I earn a small commission for each purchase at no additional cost to you.