Watching o3 guess a photo's location is surreal, dystopian and entertaining

Post by charcircuit » Sat, Apr 26, 2025, 8:01 PM UTC

cco wrote: post #187

joenot443 wrote: post #122

Earlier quoted context omitted.

Super cool, man. Watching pro Geoguessr is my latest break-time activity, these geo-gods never cease to impress me. One thing I'm curious about - in high level play, how much of the meta involves knowing characteristics about the photography/equipment/etc. that Google used when they shot it? Frequently I'll watch rainbolt immediately know an African country from nothing but the road, is there something I'm missing?

Meh, meta is so boring and uninteresting to me personally. Knowing you're in Kenya because of the snorkel, that's just simple memorization. Pick up on geography, architecture, language, sun and street position; that's what I love. It's clearly necessary to compete at the high level though.

Post by fundingshovel » Sat, Apr 26, 2025, 8:04 PM UTC

brundolf wrote: post #199

fire_lake wrote: post #177

Earlier quoted context omitted.

Which also fits with how it performs at software engineering (in my experience). Great at boilerplate code, tests, simple tutorials, common puzzles but bad at novel and complex things.

Yep. But wonderful at aggregating details from twelve different man pages to write a shell script I didn't even know was possible to write using the system utils

Post by neves » Sat, Apr 26, 2025, 8:10 PM UTC

wongarsu wrote: post #179

thegeomaster wrote: post #39

For all of the images I've tried, the base model (e.g. 4o) already has a ~95% accurate idea of where the photo is, and then o3 does so much tool use only to confirm its intuition from the base model and slightly narrow down. For OP's initial image, 4o in fact provides a more accurate initial guess of Carmel-by-the-Sea (d=~100mi The clue is in the CoT - you can briefly see the almost correct location as the very first…

4o is already really good. For most of the pictures I tried they gave comparable results. However for one image 4o was only able to narrow it down the the country level (even with your CoT prompt it listed three plausible countries) while o3 was able to narrow it down to the correct area in the correct city, being off by only about 500m. That's an impressive jump

Post by throwaway314155 » Sat, Apr 26, 2025, 8:21 PM UTC

kqr wrote: post #249

simonw wrote: post #152

Earlier quoted context omitted.

Not at all. Models have no invisible internal state that they can access between prompts. If you ask "how did you know that?" you are effectively asking "given the previous transcript of our conversation, come up with a convincing rationale for what you just said".

On the other hand, since they "think in writing" they also do not keep any reasoning secret from us. Whatever they actually did is based on past transcript plus training.

Post by jdiff » Sat, Apr 26, 2025, 8:24 PM UTC

fire_lake wrote: post #177

brundolf wrote: post #165

Earlier quoted context omitted.

I find this type of problem is what current AI is best at: where the actual logic isn't very hard, but it requires pulling together and assimilating a huge amount of fuzzy, known information from various sources They are, after all, information-digesters

Which also fits with how it performs at software engineering (in my experience). Great at boilerplate code, tests, simple tutorials, common puzzles but bad at novel and complex things.