Post
OK, this is genuinely scary. It's a picture of me, aged 5. It was taken in 1960. It was scanned and uploaded to Google Photos in 2002. No one has told Google where it was taken. But the map, bottom right, is correct.
Obviously, in 1960, my father's pre-war Exakta camera did not have GPS. Could you identify, from that picture, exactly where it was taken?
Google can.
@simon_brooke that doesn't seem to me like something Google can do, and I don't see it happening with any of my old photos.
what happens if you try to edit the location in Google Photos?
if you download the original, is there anything in the image metadata?
@simon_brooke Yes, it is scary - because it's a clear demonstration that They™️ don't have to surveil you, personally & directly, just as long as they surveil enough people to extrapolate information about you from the overlaps.
To a machine which can compare millions of images taken in the same sort of location & look for similarities, geo-locating a seemingly featureless picture like this is fairly trivial; it's just surveillance + brute-force computing.
i did a google of @TheBreadmonkey's holiday picture, and it gave me the right location and possible hotel it was taken from.
The total amount of global AI compute power in 2025 is estimated to be approximately 79 million NVIDIA H100 GPU equivalents across the top ten nations, with the U.S. having the largest share. The overall power required to run this capacity is estimated at around 55 gigawatts (GW).
https://www.mescomputing.com/news/ai/the-top-10-most-ai-dominant-countries-in-2025-report
New York City: Averages 5,500 MW
gw=1e9
mw=1e6
55e9/55e8=10 New York Cities
The NVIDIA H100 GPU offers immense compute power, reaching up to ~1 PetaFLOPS
FLOPS = Floating Point Operations Per Second
It is a meaningful way to measure performance in CPU intensive operations, like computational fluid dynamics. It's still used some in the super-computing benchmark arena.
It's not a meaningful measure of personal computing performance because most operations in your PC are not complex calculations. Much of the PC's work is moving data around or operating system activities that are short on complex matrix math.
peta=1e15
1e15*8e7=8e22 Floating Point Operations Per Second
@amiserabilist @simon_brooke @TheBreadmonkey
Finding where a picture was taken is one thing. LLM slop machines are another. What a waste of resources
Yeah also data bus speed is a big factor
@simon_brooke A.I. is being used to do a lot more than write last-minute book reports.
@simon_brooke There's a couple of things that strike me as odd about this. First, Google Photos didn't exist in 2002 it came out in 2015 so it couldn't have been uploaded to it in 2002. Second, on my account whenever Google tries to tag a location based not on the EXIF it says "Estimated Location" and yours doesn't.
Is it possible this was geotagged in Picasa or some other photo software and then uploaded to Google Photos later? I'd download the photo and see if the location is in the EXIF.
@KalenXI I can't tell you that you're wrong, since I don't remember precisely. Google Photos tells me a file date, which I assumed was the date they were uploaded, but may be the date they were scanned.
@simon_brooke Likely the date they were scanned. I notice also that the edit location pencil icon is greyed out. At least on my photos that only happens if a photo is uploaded with a location already embedded in the metadata. So it's likely that photo had the location added before it was uploaded to Google Photos.
@simon_brooke Have you ever seen videos of Geoguessrs? They cann tell within seconds where they are on Google Streetmap. They recognize the patterns. That is what machine learning ist best at.
@simon_brooke The place Google knows as Carrick Beach does not have buildings in backlground. (photo 1)
Carrick Bay does. (the two are close to each other). The houses appear to be in town/township named Castle Douglas, the beach next door is in town named Borgue. (asscoding to Google Maps).
This leads me to beleive that Google didnt geolocate your image but rather found a similar image which had been tagged with "Carrick Beach" in someone's personal album. Likely combined with context of your Internet presence pointing to Scotland.
Consider as well that if photo is from 1960, the buildings may have changed (and more built) since then, so that makes placing image even harder. Google Image search for Carrick Beach or Bay did not yield similar images, so I have to assume it drew from scrouring other persons personal albums.
@simon_brooke so I’m not a great geoguesser but there are people that are scarily good.
Looking at this pic I see a row of cars, that might give an idea of the country. It’s likely a beach but it’s not purely sand so geologically that may limit it further.
These are the kind of things that help surprisingly.
But Google has access to all the data of Google Earth & some AI so that’s my uneducated guess. Also if you uploaded it they’ve all the data they’ve collected on you over the years to go on.
@simon_brooke This works quite reliably with interiors of restaurants or conference rooms (whenever someone has posted a photo online, such as on restaurant websites), as well as tiny parts of photos (e.g., a tiny section of a lifebuoy on a well-known boat) and the location (which harbor or coast). Google also recognizes people from behind based on their clothing and other people and sometimes tags them (if accounts are linked).
@simon_brooke
Yep it's a known fact now that AI has gotten amazingly good at getting the location of an image based on the environment, sky, landmarks etc.
@simon_brooke comparison to many othr similar photos which are tagged
@simon_brooke amusingly I tried a Google image search of a couple of selected features...
The collection of stones to the left of the photo, their "AI Overview" says: "The image is part of the extensive data collected by Mars exploration missions designed to study the planet's geology and search for signs of ancient water."
I told it: "This is a photo from a beach on the planet Earth, do you know where the beach is."
It refuses to believe me:
"This image is not from a beach on Earth; it is a photograph of the Martian surface, specifically from a NASA rover mission. Visual search results consistently match the image with photos identified as taken by Mars rovers, such as Spirit or Curiosity, rather than any specific location on Earth."
...
Would you like me to find out which specific NASA mission captured this image?
So well done on your 1960s holiday on Mars! AI says so, it is irrefutable.
@ai6yr I believe this is up your alley 😉
@simon_brooke if I knew the location, yes, I'd be able to say where it was taken. You know that annoying person who tells what films actors have been in while you're watching a film? That's me, but add in locations, including inside buildings.
Google "knows" the surrounding scenery because it will have other people's geotagged images in its data banks.
I took a photo of a Star Trek book cover once, and Google tagged the scenery in the background correctly 🫠
It's a whole class of machine learning model. SotA a few years ago was galled PIDGEOTTO.
@simon_brooke apparently just these two things..
@simon_brooke I bet some of your father's other relatives out friends has uploaded the photograph to Google and manually tagged it correctly.
When you uploaded your copy, Google recognized it's the same picture.
@simon_brooke
For good or ill: "Google’s mission is to organize the world's information and make it universally accessible and useful."
https://www.google.com/intl/en_us/search/howsearchworks/our-approach/
@simon_brooke no..but the waters really cold!
@simon_brooke wow this IS scary. Is it possible it was uploaded on that location and google assumed upload as foto location?
@Ilka4You no. It wasn't. Uploading pictures, in 1960, wasn't a thing.
@simon_brooke I onviously meant the uploading location at the time it was in fact uploaded.
@Ilka4You yes, but that was about twenty five miles from the place in the picture. That might be near enough to narrow the search, I suppose, but astonishing that anyone would think it worthwhile.
@simon_brooke google doesn't think, it compiles data. If you upload a bunch of fotos from 1 day, 1 location, 1 camera, google finds similarities, even if you do not provide any of this info, it compares it to the info it already has. That is why I asked if the picture was uploaded from the location - it is the easiest starting to make out the location in the picture by comparison of the surrounding area. The photo quality can help identifying the actual time frame.
@Ilka4You The camera the first photograph was taken with was manufactured before 1939; the photograph was taken in 1960. It was scanned (along with hundreds of other photos from the family album) after my mother's death in 2001, and uploaded to Google Photos sometime since then.
I believe there was no metadata in the files that were uploaded; I will check tomorrow.
There is one other photo in the set taken with the same camera on (I think) the same day; it is also geolocated to the same place.
@simon_brooke this is scary indeed!!😱
@simon_brooke rainbolt in shambles
@simon_brooke It is called data correlation. They have enough cartographical datas in their databases to pinpoint almost any location shown on the photos. Geography, topography, cartography, the location of buildings and their plots, streets/roads network developed over the time, land registration numbers, and so on. One thing that must be done is to simply connect these items appropriately. And that's it. People can not hide themselves anymore. And that's most scariest thing. There's no privacy
@simon_brooke According to Wikipedia, it works by identifying major landmarks or landscapes.
Interestingly enough I looked at your photo without noticing the map and thought "that looks a bit like Carrick beach, but could be loads of places probably" So it is not a million miles beyond human capability, if like me you live near the place and have been there lots.
@simon_brooke I’m torn between horror and respectful amazement. How indeed??? At first I thought, well, they probably matched up the houses in the background, but then I looked closer and saw that those are cars! And the pattern of rocks could probably be matched, given enough data, but they’re so small that I’m sure they must have gotten moved around over the years. And, really, what else is even in this photo?
@simon_brooke be careful now with any images of kids. Google flags things and people are being dragged into shit unnecessarily. Do not store images longterm with their or any of the other main cloud services.
@simon_brooke Google street view, modern GPSed photos and GeoGuesser are a very rich data set 🤔
@simon_brooke Their ability to pinpoint an image’s location from minimal clues is going to be very handy for police states. And with Google’s new motto apparently being “Be Evil,” you know they will not miss an opportunity to make a buck by handing over the locations of dissidents.
@simon_brooke Google Lens has been able to geolocate old photographs for a few years now. This article from three years ago gives an indication of some of their information sources from that period. Google will have vastly more information now, including other old photos, contemporary photos (Google Maps Street View includes 360° images from hikers), old maps, satellite photos, etc. The amount of data they have must be truly mind-boggling. Combined? Very powerful.
https://www.bellingcat.com/resources/2022/08/09/using-new-tech-to-investigate-old-photographs/
@simon_brooke I bet some experienced geoguesser players could 
@simon_brooke
That is all scary indeed.
This is genuinely astonishing and scary. I will try this with some scans of old photos, I have lying around.
@simon_brooke Geolocation of photos by visible elements on the background has been a thing for a while, that's why e.g. military photos undergo strict check by specialists, eventually background blurring, before being released, and failure to do so really costs lives.
It is because of AI. It obviously recognizes location according to landscape. You really cannot escape from their AI bullshit anymore. Feels like privacy intrusion.
@simon_brooke Google can reverse engineer images to find matches. If a similar background photo is available online it's possible for Google to match it.
Friends have done similar matches in identifying social media trolls from countryside snapshots used in profile pictures.
@simon_brooke Our local cycling community has this game where you should identify where a photo of a bike with background is taken, cycle there, take a pic and post a new pic. You'd me amazed at how good people are identifying locations, but they are taken now, not tens of years ago and the most frightening is that a computer can do it.
@simon_brooke Scary, indeed. Imagine, a fascist system using #Google.
BTW, they don't need GPS data from a camera. The have the largest photo data base with street/landscape photos, trained #AI #machineLearning to identify structures and compare #photos. If trees or walls are still there, it's easy. They used users for trainings like "click all parts with a bus". And they have your faces for #faceRecognition in the future ...
Eerie… but then again context is everything. Google has access to a huge amount of information in the images and exif information if available. Correlating all of this across its huge user base provides possibilities we cannot even imagine.
These companies and their tools already know more of us than we know about ourselves. We are the product.
Ever realized why we need rules and regulations around privacy?
@simon_brooke Well, they have cars driving around taking 360° pictures of every place humans go to, and they're pretty good at searching, so it's not that surprising that they can do that. But it is scary, indeed.
If that isn't startling enough, try this one. This one shows a cat crossing a dirt track, with a dry-stone dike behind it. Above the dike, trees of a small spruce plantation can be seen.
The picture was taken in about 1966. I think by me with a box brownie -- a very cheap, primitive camera.
Again, the location shown is correct within 50 metres. HOW?
@simon_brooke Is it possible that there were keywords hinting the location in the original uploadwd file? Maybe even folder names? Do you have a copy of the file that was never online? Would be interesting to check that. Thanks for sharing this!
@animaux No, it's not possible that there was anything hinting at location in metadata, the images were just scanned from a photograph album. Nothing clever was done with them.
Could someone in the family have geolocated them since? That is just possible, but I wouldn't think so. And it wouldn't account for a photo of the back of a family house in Kirkcudbright, which is geolocated to the middle of the Irish Sea!
@simon_brooke with data and inference a lot is possible, i'm mostly surprised that they're allocating resources to do it automatically
could be interesting to upload pictures of places that changed dramatically since the 60. I know a lot of fields are now urban projects now. can google trim its reasoning on data set from a particular decade ?
@jnpn They do their machine leraning also with historical photos. I once did a research to identify a small road in the USA with a photo of the beginning 1950s. I landed on streetview of nowadays and they had even marked the empty places of houses (one on my photo) that were demolished in the 1980s.
@NatureMC @simon_brooke so much of their operations that we don't know of ..
@simon_brooke I wonder if other people have geotagged similar (enough) pictures of the same place.
Are there any that are obviously wrong? Or just some that don’t get geolocated?
@drnoble there are one or two that are very wrong. But not many.
@simon_brooke interesting, so not just lucky guesses.
@simon_brooke I see a telegraph pole with wires crossing a track. Is it still a dirt track?
Plus what I thought was another telegraph pole, maybe an end pole, near the left of the photo, although I don't see cables from there to the central pole.
Second guess: a pub sign, no, it's something with a junction box on it. Utilities?
@grahamperrin It is still a dirt track (and the dike is still there), but the electricity poles are now different, and the spruces in that wood have been felled (although there is still a wood there).
This is where Google thinks the photo was taken. The actual location was outside the white cottage about 30 metres north of the location Google has guessed.
Interestingly, this picture of my little sister and her cat -- taken only a few minutes before or after the other cat photo, with the same camera, on the same roll of film -- is not geolocated. There's nothing visible over the dike in this one.
Shot in the dark: I assume that these images were uploaded to a website in 2002 (blog, forum, etc). Then as you said at some point since then Google Image search indexed them and made them available to find when searching.
Is there any additional contextual info on the website they were uploaded to? Like if they were in a blog post titled "Vacation at Such-and-Such Beach"?
With that little info it's often enough (which is scary). But if there is nothing then it's even scarier.
@simon_brooke They probably ran a cross-comparison with their google map/earth data.
Your first two photos showed distinctive landmarks that probably remain unchanged up till this day; your third photo didn't.
Based on this theory, if you have a photo that was taken in front of an industrial estate that no longer exist today, it will unlikely be able to geolocate that.
Someone with more extensive OSINT experience might be able to explain better.
@simon_brooke I tried this once, with the paid model no less.
It thought we were on a completely different hemisphere.
@simon_brooke i wonder if the telegraph poll helps, in the UK every single one of them has a unique number so the location is known.
Or algorithmic calculation using your meta familial data - publicly available - and their Google earth data. Equally scary.
@barfilfarm Well, presumably. The cat photo is just outside the cottage we had in the 1960s... but where did Google get that information? It could be recognised in other photos taken at the same time, but there's nothing of the cottage in the picture.
Certainly they're mashing together huge amounts of data -- but I had not conceived that it would be economically viable to do that for random photos taken by unimportant people.
@simon_brooke
Could facial recognition coupled with property ownership databases be enough to link it?
@barfilfarm
@simon_brooke every photo helps to train their machine learning. The more, the better. @barfilfarm
@simon_brooke
Maybe they trained their AIs long enough by letting people play geoguesser?
@Hippie that's extremely likely. I'm quite certain that I have seen the exact same cat at least twice in the McDonald's searches. I had to travel miles.
@simon_brooke
Is it using the upload location?
@Alternatecelt I scanned them all when I was living in Auchencairn, and almost certainly uploaded them all from there. The beach photo is from Carrick shore, by Gatehouse. The cat photo was taken on a farm about two miles from Auchencairn, so fairly local to where it was uploaded.
Pictures from Islay and from Skye in the same set are not geolocated despite having far more identifiable detail.
@simon_brooke @Alternatecelt Can you rule out that they were located from associated text? You have precisely located them in your posts here; presumably you have not mentioned them online previously?
@kbm0 @Alternatecelt there is no associated text. Just, after my mother died, I scanned all the pictures in her family album and uploaded them to Google Photos for the rest of the family to see.
@simon_brooke
The rest of the family might be the key here?
@kbm0 @Alternatecelt
@simon_brooke @kbm0 @Alternatecelt They also used the pictures and location data’s from your phone and the phones of your relatives and friends (contacts).
@simon_brooke @Alternatecelt I wasn't thinking about directly associated text, more likely that you or someone else had discussed the photos in a social media thread or similar. That would be enough to pinpoint them.
@simon_brooke How's it done that, I wonder? Very unpleasant
@janeishly They do have a huge library of pictures; presumably they can compare them. And the beach photo -- well, there's not a lot of detail in that photo, but nevertheless thousands of people must have taken pictures of that beach.
But the cat on the track? That's on a farm the present owner of which is very hostile to strangers. It's *really* unlikely that there are many photos of it.
@simon_brooke @janeishly Correlation with StreetMap photos?
BT Free is a non-profit organization founded by @ozoned@btfree.social . It's goal is for digital privacy rights, advocacy and consulting. This goal will be attained by hosting open platforms to allow others to seamlessly join the Fediverse on moderated instances or by helping others join the Fediverse.
