LLMs basically are fancy calculators that spew bullshit meant to make it seem like it knows what it's talking about. The bubble is already showing its weakness, Open AI is going to implode.

Being that there is no dialogue, it doesn't need closed captions. What you might be asking about is subtitles (text of dialogue and description of audio/music) or audio description (a separate audio track which speaks the text and describes the action).

I'll still look at the evaluation objectively, and won't base my overall opinion on the lack of following instructions... But come on...

What looks even worse? A VPAT that isn't completed properly.
1. Remove the instructions pages
2. Fill out the names and dates

I'm going through one right now that doesn't actually say what product it is evaluating... It could be anything.

Very good information. It should be shared with the youth robot clubs.

Maybe a tarp is the next thing to try 😅

I don't think we read the same book...

Our Lady of Divine Computation.

My theories are:
A. NASA has been gutted by the administration's war on Federal spending so the comms department is silent, or...
B. They have nothing to report other than, "huh, that's weird and interesting..."
Two very Occam's Razor theories, but it applies right now.

So the reason is they were doing an expensive cosplay?

You again?

I eat at least 3 a day, it helps my ulcers thrive.

Thanks for spreading the word, Bill!