←back to thread

Which year: guess which year each photo was taken

(whichyr.com)
712 points trymas | 1 comments | 17 Apr 25 10:42 UTC | HN request time: 0.198s | source
Show context
peripitea ◴[20 Apr 25 06:53 UTC] No.43742005[source]▶
Tried plugging it into the new o3 model out of curiosity and it got 0.2, i.e. it got everything exactly right save for one that it missed by a year.
replies(1): >>43742292 #
delusional ◴[20 Apr 25 08:03 UTC] No.43742292[source]▶
You know, there are easier ways to cheat at the game? The photos are from the internet, a reverse image search will surface all of them.
replies(2): >>43742325 #>>43742385 #
mr_mitm ◴[20 Apr 25 08:12 UTC] No.43742325[source]▶
I read it as more of an experiment to gauge the capabilities of AI than an attempt at cheating.

I'm impressed that AI is that good.

replies(1): >>43742413 #
codetrotter ◴[20 Apr 25 08:33 UTC] No.43742413[source]▶
Is it good, or are those photos already in the training data that the model was trained on?

Would be interesting to see how well it would fare on five photos from different years, using pictures that have never previously been uploaded anywhere.

replies(2): >>43747296 #>>43747455 #
1. fooker ◴[20 Apr 25 23:57 UTC] No.43747455[source]▶
> are those photos already in the training data that the model was trained on?

The photos are of course in the training data. Most humans too, would have seen a large fraction of these photos before.

But associating it with a year is where the magic happens.

You are dismissing a technology capability that was science fiction a year or two ago.