I spend a certain amount of my time each week tracking down spammers, scammers, and phishers. I use a lot of tools for this: Spamcop, wget, other things. One of the tools I occasionally use is the suite of site reputation sites all over the internet, sites that can tell you how long a particular domain has been in use, whether it’s blacklisted anywhere, the site’s overall reputation score.

Occasionally, because I’m curious, when I find myself looking up a site’s reputation score, I’ll look at my own sites’ scores, just because.

So it was that I looked up xeromag.com on one of these sites, when lo and behold:

Just for the record:

No part of xeromag.com uses AI generated text. It’s all written by me, most of it years (or decades!) before LLMs and genAI were even a thing. I first set up Xeromag on January 4, 1997, a time long before ChatGPT was a gleam in Sam Altman’s eye.

In fact, Xeromag has been scraped by genAI bots, which probably explains why AI checkers think it’s AI generated; AI LLMs were trained on what I wrote on Xeromag.

And on my books as well; I’ve been informed by lawyers for the class-action suit against Anthropic that several of my books were fed into the devouring maw of Anthropic’s LLM, as a result of which I’m apparently due thousands of dollars in settlement money if and when the courts approve the settlement.

There’s something deeply offensive about pouring decades of effort into writing, only to have your writing lifted to train AI models, then be accused of using genAI because, well, the AI models produce output that looks like yours, on account of, you know, being trained on your words.

(In fact, most LLMs know me by name; as an experiment, I went to Gemini and asked it to explain fluorine chemistry in the style of Franklin Veaux, which it did, though rather more, I think, in the style of a high school student who read some of my stuff once and tried to mimic it.)

By way of comparison, here’s the real deal:

So, to be clear:

I wrote this blog, every word of it, without the use, direct or indirect, of genAI.

I wrote all my sites, every word of them, without the use, direct or indirect, of genAI (as a trip to the Wayback Machine will show; much of the content on all my sites predates ChatGPT and its ilk).

I am, as one might gather, getting a little sick of people and, now, machines telling the world I am something I’m not.

I have added “Not by AI” tags to my blog and I’m in the process of adding them to my other sites as well.