Artificial Intelligence

I haven’t used Claude Fable at all since it came out. Mostly I’ve been intimidated by the credit usage and feeling like I don’t really have anything that’s “Fable-worthy” (whatever that means). But it’s been a while since I ran a Lovecraft test and I decided to pit what’s maybe? arguably? the world’s best LLM […]

I spent the weekend chasing a rabbit down a hole. I built an IMDB for actual plays…mostly from my phone. The thought process went something like this: Aabriya Iyengar has been in a bunch of live TTRPG actual plays, both the mainstream kind (Dimension 20, Critical Role, The Adventure Zone, Dungeons & Dragons official) and […]

I saw a post on LinkedIn the other day from a self-proclaimed 20 year agency veteran of WordPress saying that was it, they’re moving the entire agency off of WordPress and onto AI. Now, because I, too, am a 20 year veteran of WordPress, this kind of story catches my attention. He posted that they […]

This is based on a thread that I posted to Bluesky that I decided to keep here for posterity. If you want to discuss on Bluesky, follow me at @jazzsequence.com and let’s chat. Let’s talk about AI. I’ve been using it a lot recently. It’s not that I don’t see or care about the environmental […]

For the last couple weeks, I’ve been building a headless Next.js frontend for this site — a project I’ve been calling jazz-nextjs. The idea is straightforward enough: keep WordPress as the content management layer (where I actually like writing) while serving the public-facing site through a modern React frontend hosted on Pantheon’s Next.js infrastructure. What’s […]

I want to talk today about Sora 2 and the new Sora AI app. First of all, Sora is not new. OpenAI released Sora about a year ago initially and, at the time, it was a sort of hidden part of ChatGPT that let you make bad videos. And trust me, they were bad. I […]

At the same time as I was fighting with GPT-5 to give me the thing I actually wanted, I also used the Gemini app and the access I have to pro-level Gemini models to do the Lovecraft test with 2.5-pro. Previously, I had only tested with Flash-2.0, so, not only should this be a more […]

GPT-5 is out now and I’ve only had a couple occasions to interact with it directly. So, I thought it was time to run it through the Lovecraft test. ChatGPT has adopted this quirk of late where, when it’s done answering your question or responding to your prompt, it says something to the effect of […]

I’ve been sitting on this one for longer than I should have (and then, obviously, forgetting entirely about it). After putting DeepSeek through the ropes for my Lovecraft test, I decided to try Google’s Gemini. I haven’t used Gemini much, although I have used NotebookLLM which is based on Gemini, and I think it generally […]

The stupidly big news in LLMs right now is that DeepSeek, a LLM built by a small Chinese team that was released a couple weeks ago, performs as well as, if not better than larger, more sophisticated (at least in terms of the dollars and hardware thrown at it) models by OpenAI and Anthropic (Claude) […]

I’ve run this test twice already and I figured it was probably about time to try it again. If you’ve been reading my AI posts (which is all I post anymore), you may remember past incarnations of what I like to call The Lovecraft Test. I ran it very early on with ChatGPT when it […]

Historically, prompting ChatGPT (or DALL-E through ChatGPT) for variations or subtle modifications of a single image results in a sort of image degeneration. Not only does the image not get reproduced with the slight variations, but it gets further and further away from what you originally requested. This is something that has been noted on […]