Artificial Intelligence
-
Waxing Desperate with Imagination*
Generative AI isn’t reliable. Like an unreliable narrator in fiction, you have to stay alert; what an LLM tells you only might be true. If you’re writing fiction you might give the reader a hint that your narrator can’t be trusted. You might introduce them as a court jester; a clown. Or you might design… Continue reading
-
The great human document
Another idea from Jaron Lanier: large language model software systems are not artificial intelligence. A large language model is one enormous document comprised of contributions from countless people. It’s “something like a version of Wikipedia that includes much more data, mashed together using statistics.” It “can be thought of as illuminating previously hidden concordances between… Continue reading
-
Data Dignity and Xanadu
Jaron Lanier is always interesting to listen to or read. Just this week he was a guest on Neil Degrasse Tyson’s Startalk. The discussion turned to AI (in its current form), and Lanier brought up his notion of data dignity. This hasn’t gotten much attention, as far as I can tell. According to Lanier, the… Continue reading
-
Oh spare us. No, really, I mean it.
I just found Dario Amodei’s apologia The Adolescence of Technology, from back in January. It’s quite long; Claude must have used tons of tokens writing it for him. Amodei can, of course, afford it. His essay is chock full of hand waving dismissals of reservations reasonable people might have about large language models (and the… Continue reading
-
When you’re paid to undermine yourself
Minas Karamanis has done some thinking about using LLMs. How they can help and, on the other hand, how they can undermine your own abilities and career. This is probably obvious, but it all depends on how you use them. Karamanis is an astrophysicist at Berkeley, but I think what he says applies to any… Continue reading
-
This is what’s wrong with US business models
Chinese large language models like DeepSeek are smaller and more efficient than the centralized commercial behemoths from OpenAI and Anthropic, et al. They’re also open source, so you can use and implement them however you want. They’ve been released to the world and they’re free and open. The US response is to create a new… Continue reading
-
All too human
Sam Altman, the public face of OpenAI (ChatGPT vendor) is most of all a fast-talking salesman. He’s really good at that. As such, you can’t really believe anything he ever says; when he says “A”, it’s just a reason to obtain “B”, no matter what it takes. Truth is entirely beside the point. Seems like… Continue reading
-
Questions unasked
There was a Cisco AI Summit event yesterday. It was a corporate-captured attempt at a TED conference. They did, to their credit, attract the top names in AI, including Sam Altman (of OpenAI) and Jensen Huang (of Nvidia). The whole point, if there was one beyond real-time hagiography, was to get the tech celebs to… Continue reading
-
Personal Software
AI coding assistance makes it possible (even relatively easy) to create applications for your own personal use. They can be weird, quirky, and “missing features.” It won’t have all those dozens of menu items, hundreds of icons, and thousands of features that commercial software contains in order to cater to every possible customer. Because there’s… Continue reading
-
Oh, right, “AI”
“…the world’s shadiest art dealer… is pretty sure that if we keep shoveling words into the word-guessing program it will wake up and become intelligent. Which is just, you know, stupid. It’s like thinking that if we just keep breeding our horses to run faster, one of our mares will eventually give birth to a locomotive.“… Continue reading
About Me
I’m Pete Harbeson, a writer (among other things) located near Boston, Massachusetts. In addition to writing my own content, I’ve learned to translate for my loquacious and opinionated pup Chocolate Bossypaws. No surprise, she mostly speaks in doggerel. You can find her contributions tagged with Chocolatiana.
Check out my other blog, Techlimitics, where I’m grappling with the nature of simplicity. You can also find some of my minor software projects at GitHub. Nothing very impressive. I mostly write tiny utilities in Python.
I find myself suddenly de-corporatized (their choice, not mine). To help keep the lights on, buy me a coffee!