As some may have noticed I am on a publishing/speaking hiatus (no posts since June 2025 and I generally used to do at least one a month). I only do some commenting, currently. But this I really have to share.
Today I found out one of my key articles (On the Psychology of Architecture and the Architecture of Psychology) had been used by a GenAI setup as a basis to create a podcast episode on YouTube. A youTube channel, NLMDD — for Notebook Language Model Deep Dive — created a GenAI generated podcast discussion video based on that article. The script is LLM-generated, then two (probably artificial) voices do the talking. There is no moving imagery. It could be a near automated production, throw in a link and out comes a podcast episode.
This one seems to start out reasonably well (though, tone and wording: ugh, beige and sycophantically positive about my post), but then it derails, saying stuff that isn’t in my article at all, sometimes even the complete opposite of my own thoughts, more typical self-help guru talk straight from the internet. Jay..
That is not unexpected. Apart from typical GenAI-unpredictability — which can be more or less an issue — Generative AI reworking (such as a summary) of input tends to show — as I wrote earlier — a tug of war between the content (‘context’ in GenAI terms) on the one hand and the training material (via the parameters) on the other (see When ChatGPT summarises, it actually does nothing of the kind). So, the video is called a ‘deep dive’, but what it is is `AI-slop’, sorry. Luckily so far, the video only has a single view: mine 😎.
It is clear that I Iike being part of a podcast (all writers/speakers are addicted to attention/valuation), but I am not happy that my stuff gets mangled into this junk. People are much better off reading the real thing. Takes about the same amount of time. Or watch my Complexity Crunch video (link starts 35min in, the segment takes 10 minutes, which is about half of that AI generated podcast AI-slopcast.
Note to you: algorithmic ad placements make sure that ads are placed in such slop and every time someone sees that, the advertiser — any reputable brand — pays money to Google and Google pays a bit to the channel. So, do you really want to watch it? This is how click-bait misinformation from the worst kind is funded by reputable businesses. Hey! Reputable Business! Are you certain your ads don’t fund AI-slop driven misinformation, extremism, etc? No you are not. Well, thank you very much for sponsoring the death of the civil society.
[You do not have my permission to use any content on this site for training a Generative AI (or any comparable use), unless you can guarantee your system never misrepresents my content and provides a proper reference (URL) to the original in its output. If you want to use it in any other way, you need my explicit permission]
R&A IT Strategy & Architecture 