Comment by stingrae
1 day ago
1 and 2 have been achieved.
4 is close, the interface needs some work to allow nontechnical people use it. (claude code)
1 day ago
1 and 2 have been achieved.
4 is close, the interface needs some work to allow nontechnical people use it. (claude code)
I strongly disagree. I’ve yet to find an AI that can reliably summarise emails, let alone understand nuance or sarcasm. And I just asked ChatGPT 5.2 to describe an Instagram image. It didn’t even get the easily OCR-able text correct. Plus it completely failed to mention anything sports or stadium related. But it was looking at a cliche baseball photo taken by an fan inside the stadium.
I have had ChatGPT read text in an image, give me a 100% accurate result, and then claim not to have the ability and to have guessed the previous result when I ask it to do it again.
>let alone understand nuance or sarcasm
I'm still trying to find humans that do this reliably too.
To add on, 5.2 seems to be kind of lazy when reading text in images by default. Feeding it an image it may give the first word or so. But coming back with a prompt 'read all the text in the image' makes it do a better job.
With one in particular that I tested I thought it was hallucinating some of the words, but there was a picture in the picture with small words it saw I missed the first time.
I think a lot of AI capabilities are kind of munged to end users because they limit how much GPU is used.
[dead]
I dispute 1 & 2 more than 4.
1) Is it actually watching a movie frame by frame or just searching about it and then giving you the answer?
2) Again can it handle very long novels, context windows are limited and it can easily miss something. Where is the proof for this?
4 is probably solved
4) This is more on predictor because this is easy to game. you can create some gibberish code with LLM today that is 10k lines long without issues. Even a non-technical user can do
I think all of those are terrible indicators, 1 and 2 for example only measure how well LLMs can handle long context sizes.
If a movie or novel is famous the training data is already full of commentary and interpretations of them.
If its something not in the training data, well I don't know many movies or books that use only motives that no other piece of content before them used, so interpreting based on what is similar in the training data still produces good results.
EDIT: With 1 I meant using a transcript of the Audio Description of the movie. If he really meant watch a movie I'd say thats even sillier because well of course we could get another Agent to first generate the Audio Description, which definitely is possible currently.
Just yesterday I saw an article about a police station's AI body cam summarizer mistakenly claim that a police officer turned into a frog during a call. What actually happened was that the cartoon "princess and the frog" was playing in the background.
Sure, another model might have gotten it right, but I think the prediction was made less in the sense of "this will happen at least once" and more of "this will not be an uncommon capability".
When the quality is this low (or variable depending on model) I'm not too sure I'd qualify it as a larger issue than mere context size.
1 reply →