I'm a pretty avid member of various history groups, and one thing that has absolutely driven me nuts for the past couple of years is how many people there are that use AI for upscaling and colorization of photos - not knowing or noticing how the models fundamentally alter the photos. A couple of zooms in on the photo, and it is nightmare fuel.
A week ago me and some members spent a couple of hours trying to find a building from the early 1900s, because someone had uploaded a photo and asked about the building. Sifted through old maps, newspapers, etc. but couldn't find anything. Turns out said photo had been upscaled via AI, which in turn had added some buildings here and there.
But, yeah, for stuff like OP posted it could work out nicely.
One random example of a before/after: https://imgur.com/a/WIAYLHm
But the majority of the commnents (including the top comment) on this thread are about how bad AI Images are and how bad AI is in general, how it is altering history etc -when the author didn't even do any of that in his post
It shows the mindset of the community these days more so than the technology.
Ideally they'd always carry an "AI-generated" flag (in the db and in the frontend) until manually reviewed (or never) by a human. If anything, this is actually in AI proponent's favor as it would let you periodically regenerate or cross-validate (a subset of) the AI contributions some years down the line when newer and better models are released!
https://www.oldnyc.org/#707133f-a this is supposed to be here https://www.oldnyc.org/#702487f-a
also, if folks are interested in these old depictions of NYC, check out https://1940s.nyc/ as well!
But it's actually really cool how they used AI to better determine the locations of the photos. I love this!
I haven't seen an "AI edited" image that hasn't changed important details, and so the result is just yet more slop.
The experiance made me certain that AI is going to to much more harm than good to the buisness of archiving historical photos.
As for the lady who is distorting photos to colorize them - I don't even understand why you would want to do that. There are other ways!
If the risk of mistranslation is high, I fail to comprehend how letting AI "take a swing at it" does not reduce the translation quality?
How are they ensure no drop in translation quality?
(Provenance is so important. The infinitely-recopied local history photos were never a great source anyway).
It reminds me of the cuneiform problem. Between 500,000 and 1 million tablets have been collected. This is one of the earliest preserved writing systems. Even so, fewer than 10% of these tablets have been translated. I was surprised to learn this but it makes sense. There are several problems:
1. Scribes used a lot of shorthand;
2. Cuneiform itself changed over time;
3. Writers would use multiple languages (eg Sumerian, Akkadian), even on the same tablet. There are relatively few people fluent in these languages, particularly in multiple of them at once;
4. To some extent the tablets are 3D such that a 2D photo might not be sufficient to translate because you might need to physically turn the tablet to accurately see the marks; and
5. In some cases the tablets are incomplete or broken so you may not to figure out how things fit together.
I wonder if AI can help make inroads into this 90%. I really wonder what is waiting to be unearthed.
(I briefly got excited that there might be a street sign _in_ the photo, but if you zoom way in it says "DENTIST")
+1 to 1940s.nyc. Very different photos — those are were taken for tax assessment, the ones on OldNYC were taken to document the city as it changed. The photographer had an arrangement where he'd get tips from demolition crews, and go shoot buildings before they were gone forever.
[1]: https://digitalcollections.nypl.org/items/5a5e06a0-c539-012f...
This has been true since before LLMs, but now so many more people and use cases are enabled so much more easily. People are undisciplined and quick to take short term gains and handwave the correctness.
Its like saying "I love Da Vinci's art so I'm going to draw a moustache on everyone in the last supper" which you probably wouldn't do if you really loved Da Vinci's art.
Meh, so what if I only love Da Vinci's art to the degree that it's amusing to adulterate with mustaches?
Which is to say, I think it comes down to what you value most out of historical photos; a forensic record of truth, or general idea of what it was like to live at the time, compared to today.
Over the past two years I’ve quietly rebuilt major parts of the OldNYC photo viewer. The result: 10,000 additional historic photos on the map, more accurate locations, and a site that’s cheaper and easier to run—thanks to modern AI tools and the OpenStreetMap ecosystem.
OldNYC had about 39,000 photos in 2016. Today it has 49,000.
Most of these changes happened in 2024, but I’m only writing about them now in 2026. (I got distracted by an unrelated project.) If you haven’t visited OldNYC in a while, take a look—you might find some photos you missed.
Here are the three biggest improvements: better geolocation, dramatically improved OCR, and a switch to an open mapping stack.
OldNYC works by geocoding historical descriptions—turning text like “Broad Street, south from Wall Street” into a latitude and longitude.
Originally this mostly meant extracting cross streets from titles and sending them to the Google Maps Geocoding API. That worked well when the streets still existed—but many historical intersections don’t.
Two changes in 2024 improved this dramatically.
Some images include useful location details only in the description. I now use the OpenAI API (gpt-4o) to extract locations from that text.

Public Schools - Brooklyn - P.S. 143. 1930
Havemeyer Street, west side, from North 6th to North 7th Streets, showing Public School No. 143. The view is north from North 6th Street.
The school no longer exists, so the title alone can’t be geocoded. From the description, GPT extracted:
Both intersections exist in OpenStreetMap, so OldNYC places the image at the first one.
Tasks like this require a surprising amount of interpretation: GPT understands that “North 6th” means “North 6th Street” and extracts the relevant intersections while ignoring irrelevant phrases like “west side”. Computers have historically had trouble with this type of task, but the newer AI models nail it.
Using GPT located about 6,000 additional photos. Today OldNYC can locate roughly 87% of photos with usable location data, and about 96% of mapped images appear in the correct location.
I also replaced the Google Maps geocoder with OpenStreetMap and historical street datasets.

Brooklyn: Fulton Street – Nassau Street
These streets intersected in Brooklyn in the 1930s but no longer do today. Google geocodes this to Manhattan, where streets with those names still intersect.
OldNYC now incorporates data from the NYPL’s [historical streets project], which includes the original Brooklyn intersection. The photo now appears in the correct location.
Most OldNYC photos include descriptions from the NYPL catalog—but on the NYPL site these are scanned typewriter images, not text.

When I launched OldNYC in 2015, converting these images to text (OCR) was the hardest technical problem. I built a custom pipeline using Ocropus that achieved over 99% character accuracy. Even so, the errors were noticeable when reading.
To fix mistakes I added a “Fix Typos” feature that let users correct transcriptions. This triggered New Yorkers’ collective OCD and users submitted thousands of edits.
In 2024 I rebuilt the OCR system using gpt-4o-mini.
The results were much better:
Here’s a dramatic example where the old OCR produced complete gibberish due to an unusual font:

GPT transcribes it perfectly.
A few lessons from rebuilding the pipeline:
Overall, tools like OpenAI mean that OCR is a much easier problem in 2024 than it was in 2015.
When OldNYC launched, Google Maps was the default choice for web mapping, and it was free to use. But over time, Google’s pricing model changed. In late 2024 they replaced their $200/month free credit with separate quotas for individual APIs. Under the new system, rather than being free, OldNYC would have cost about $35/month.
Instead of paying Google indefinitely for a hobby project, I migrated the site to OpenStreetMap vector tiles and MapLibre.
The new stack has some nice benefits:
For example, I can remove anachronisms like highways and tunnels that didn’t exist in the 1930s.
Look, no Brooklyn-Battery Tunnel!
There’s still plenty to improve.
AI could extract additional information from images—identifying people, buildings, or indoor/outdoor scenes. I’d also like to incorporate photographs from other collections.
I’ve also started contributing to OpenHistoricalMap, the history-focused cousin of OpenStreetMap. If it eventually includes full historical street grids for NYC, locating photos could become dramatically easier.
Finally, I’d love to make it easier for developers to build OldNYC-style sites for other cities. If you’re interested, please reach out.
📪 If you’d like to be informed of OldNYC updates, please subscribe to the new mailing list! If you subscribed before 2026, you’ll need to subscribe again. Sorry, MailChimp deleted the old list. 😡