An update to the previous post notes that matches the actual prop: https://daringfireball.net/linked/2025/12/27/slopibus

The prop doesn't seem to have the weird intermittently blurry and inconsistently spaced maze, or at least it's not on the same side. What a weird unforced error it would be if that was the only part of the image to actually be AI-generated.

Ha, so Apple uses AI slop to make props during filming? That’s even more attention to detail than anyone ever wanted.

If you watch the show, they’re clearly intentionally meant to be off kilter.

I watched the show and I may have snoozed through most of that. It was not a good series for me

The contradictory text on the prop feels like a plausible human mistake - like from taking bits from different reference images.

Sure, but if they're going to make a holiday greeting post out of it and commission a "human" painting? It just feels off.

Follow the link on Gruber's own post. The milk label thing occurs on a prop on the actual show.

The maze has no viable path either, and odd half-drawn lines.