I’m sure there are areas where the location guessing can be scary accurate, like the article managed to guess the exact town as its backup guess.
But seeing the chain of thought, I’m confident there are many areas that it will be far less precise. Show it a picture of a trailer park somewhere in Kansas (exclude any signs with the trailer park name and location) and I’ll bet the model only manages to guess the state correctly.
Before even running this experiment, here’s your lesson learned: when the robot apocalypse happens, California is the first to be doomed. That’s the place the AI is most familiar with. Run any location experiments outside of California if you want to get an idea of how good your software performs outside of the tech bubble.
>Show it a picture of a trailer park somewhere in Kansas (exclude any signs with the trailer park name and location) and I’ll bet the model only manages to guess the state correctly.
This isn't really a criticism though. The photo needs to contain sufficient information for a guess to be possible. Photos contain a huge amount of information, much more than people realize unless they're geoguessr pros, but there isn't a guarantee that a random image of a trailer park could be pinpointed.
Even if, in theory, we mapped every inch of the earth and then checked against that data, all it would take is a team of bulldozers and that information is out of date. Maybe in the future we have constantly updated feeds of the entire planet, but... hopefully not!
I tried with various street photographs from a medium-sized German city (one of the 50 largest, but well outside the top 4). No obscure locations, all within a 15 minute walk of the city center and it got 1/7 correct. That one was scarily precise, but the other ones got various versions of "Not enough information, looks European" or in better cases "somewhere in Germany".
you never know.. LLM could go full sherlock holmes. Based on the type of grass and the direction of the wind. The type of wood work used. There could be millions of factors that it could factor in and then guess it to a t.
> Based on the type of grass and the direction of the wind.
There was a scene in High Potential (murder-of-the-week sleuth savant show) where a crime was solved by (in part) the direction the wind was blowing in a video: https://www.youtube.com/watch?v=O1ZOzck4bBI
In 2017, the Hollywood actor Shia LaBeouf (and two others artists from a trio called "LaBeouf, Rönkkö & Turner") put up a flag in an undisclosed location as part of their "HEWILLNOTDIVIDE.US" work [1].
> On March 8, 2017, the stream resumed from an "unknown location", with the artists announcing that a flag emblazoned with the words "He Will Not Divide Us" would be flown for the duration of the presidency. The camera was pointed up at the flag, set against a backdrop of nothing but sky.
[...], the flag was located by a collaboration of 4chan users, who used airplane contrails, flight tracking, celestial navigation, and other techniques to determine that it was located in Greeneville, Tennessee. In the early hours of March 10, 2017, a 4chan user took down and stole the flag, replacing it with a red 'Make America Great Again' hat and a Pepe the Frog shirt.
I just tested the model with (exif-stripped) images from Cork City, London, Ho Chi Minh City, Bangalore, and Chennai. It guessed 3/5 locations exactly, and was only off by 3kms for Cork and 10kms for Chennai (very good considering I used a slightly blurry nighttime photo).
So, even outside of California, it seems like we're not entirely safe if the robot apocalypse happens!
Context: Wisconsin, photo I took with iPhone, screenshotted so no exif
I think this thing is probably fairly comprehensive. At least here in the US. Implications to privacy and government tracking are troubling, but you have to admire the thing on its purely technical merits.
But seeing the chain of thought, I’m confident there are many areas that it will be far less precise. Show it a picture of a trailer park somewhere in Kansas (exclude any signs with the trailer park name and location) and I’ll bet the model only manages to guess the state correctly.
Before even running this experiment, here’s your lesson learned: when the robot apocalypse happens, California is the first to be doomed. That’s the place the AI is most familiar with. Run any location experiments outside of California if you want to get an idea of how good your software performs outside of the tech bubble.