Jowsey

Another Claude-written changelog. They never inspire much confidence in a project for me.

I hadn't been exposed to the new GPT-Image-2.5 model before this, so didn't have any "tells" to go off of, and found it almost impossible to begin with, but it seems the brain is pretty good at adapting for visual stuff! After 2 rounds, I was able to build up a 16-image combo, answering most within a second or two.

I think the "tell" I ended up with is that almost all of the AI images tend to center around a very obvious main "subject" (a flower, a bike, a wrench, etc) – presumably an inherent artefact of generating from a prompt – with everything else around the subject having a very strange depth to it. The depth, focus, and bokeh around the main subject just never look quite right. If you look at an image and find it has a very strong central subject with a little too much depth separation than expected, there's a good chance it's AI.

I recognise that this method of finding a consistent "tell" will probably not hold for future models, unfortunately. Everything else looks almost perfect at this point.

for those who skip to the comments: Qwen released an updated checkpoint of this model today (0902) with significantly higher benchmark results that appear to place it much closer to Fable/Sol

Interestingly, it seems almost every set of three seems to follow a pattern: one passage of the three will have a key word or phrase swapped in the first sentence. That is, for every set of 3 passages, two will start with ~identical sentences, and one will have a key word or token changed.

I caught onto this early and used it every time, and ended up getting 2/10, which is worse than random chance. I smell trickery!

So many Claude-isms in one post. Please stop.

My understanding is that these "preview" models are usually earlier RL checkpoints, and that "official release" happens when they're happy with the training run?

I believe they mentioned around the preview announcement that they'd be releasing improvements to capability, which I assume means continued training.

This is pretty standard in every model. Ask Opus or Gemini about 2026 (without a big system prompt to steer them) and they'll swear blind it's 2024/25 too.

llm/bot comment

This reply is entirely AI generated. You guys are trying to find reason in a hallucination. It's unfortunately impossible to put into words what the "LLM smell" is at this point, but I trust someone else who spends a lot of time reading LLM output can back me up on this.

I've seen these agent-written fake anecdotes on Twitter, Reddit, and now here, all with the exact same formatting. They pretend to be real people with real anecdotes, but they're all completely made up.