08 July 2025

How to Use AI to Analyze Family Photos

It's common to see people online looking for help in dating an old photo. They're looking for someone who knows about the photo's:

  • clothing styles
  • automobiles
  • uniforms
  • locations

If you ask for this type of help, your post may never find the right type of expert.

But I know where you can find the expert you need in real time. If you drag and drop a photo in a chat with Microsoft Copilot, it will give you an in-depth analysis of the photo. Copilot is available online, on a Windows PC, and on Bing.com.

Microsoft Copilot analyzes whichever photo you drop into a chat, whether you provide additional information or not.
No clue about a portrait in your family tree? Drop the file into this AI chat window for a detailed analysis.

Let's test Copilot on clothing styles, automobiles, uniforms, and locations in old photos. Copilot has different levels of response. For everyday questions, you can use Quick Response and get an instant answer. For deeper analysis, choose Think Deeper and get your detailed answer after a few seconds. If you don't see anything after you enter words or a photo, scroll down the chat a bit.

Clothing Style Recognition

Clothing Style 1. I started a chat by telling Copilot, "This woman was born in 1856 and lived in rural Southern Italy. When do you think this photo was taken?" I dropped a photo of my great grandmother into the chat. Copilot looked at her dress, her hairstyle, and the type of photo.

Based on her birth year, Copilot placed the photo in roughly 1895–1905.

Family context: To me, Marianna looks at least 65 in this photo, which would put it in 1921. Marianna would have sent this photo to her son, my Grandpa Adamo, after he returned to New York in 1920. I know it belonged to Adamo because his daughter, my aunt, gave it to me.

Copilot may have been off on this photo, but Marianna may be wearing an old dress. She didn't have enough money to be fashionable.

Clothing Style 2. My first cousin gave me her late mother's photo collection. It has photos belonging to my aunt, my grandparents, and my great grandparents.

One photo is a professional portrait of a young man. Part of its backing tore, but it seems to say the name Donato Basile. There was a Donato Basile born in 1902 in my Grandpa Pietro's hometown. He married Pietro's sister Giovannangela.

I asked Copilot, "Could this be a man born in 1902 in Southern Italy?" and dropped in the portrait. Copilot said:

Based on the styling of the suit, tie, pocket-square and painted backdrop, this portrait most likely dates to the 1920s–early 1930s. A young man in his early to late twenties of that era would often pose in exactly this fashion.

It went on to discuss features of his clothing and the portrait style. Everything points to this being a man born in 1902 in a photo taken in the late 1920s to early 1930s.

Family context: I know Giovannangela died in early 1931. Did Donato take this portrait when they married in 1924, or was this after his wife died? Since he sent this portrait to his brother-in-law in the U.S., it should date between 1924 and 1930.

You can use AI to identify objects, places, and clothing to help date a family photo.
Ask a specific question, add a little background, and this AI chat window can fill in tons of details about that family tree photo.

Automobile Recognition

Automobile 1. I have a photo of a lovely woman leaning on a car. I asked Copilot, "What type of car could this be?" and dropped in the photo. Here's what it said:

Based on the wire-spoke wheels, rounded fenders, high running boards and simple headlamps, this looks very much like a late-’20s Ford Model A (1928–31). The silhouette—sloping radiator shell, straight beltline, and absence of integrated headlights—matches Ford’s most popular second-generation car.

A search for images of a 1928 Ford Model A confirmed it.

Automobile 2. I asked Copilot, "What type of car is this?" and dropped in a photo of my grandfather Pietro beside an old-style car. Copilot said:

Based on the styling cues—open-top body, large round fenders, spoked wheels, and upright grille—this car appears to be a Ford Model T Touring Car, likely from the mid-1910s to early 1920s.

Copilot went on to describe the car's key identifying features, and it narrowed the year down to 1915–1922.

Uniform Recognition

Uniform 1. I have a photo of a distant cousin that says "Plattsburgh NY 1922". I dropped in the photo and asked Copilot, "What type of soldier is this? He was stationed at Plattsburgh Barracks in 1922." Copilot said:

Based on the uniform in the photo and the historical context, the soldier stationed at Plattsburgh Barracks in 1922 was most likely a member of the U.S. Army Infantry, possibly part of a peacetime garrison or training detachment.

Copilot gave me details about the soldier's cap, tunic, belt, and leggings. It said, "These features align with U.S. Army service uniforms from the early 1920s."

Uniform 2. My grandfather Adamo fought for the Italian Army in World War I. I have a photo of Adamo and 3 other men in their uniforms. I told Copilot, "This is my grandfather, born in 1891, and 3 other Italian men who fought in World War I. What type of service do their uniforms suggest?" Copilot recognized the uniforms and said:

All four men are wearing the standard Royal Italian Army (“Regio Esercito”) service tunic and cap of the World War I era. Their uniforms point to front-line army service—most likely infantry—rather than a specialist corps.

It went on to describe their tunics, collar patches, and shoulder straps. It also said:

Each man displays at least one medal on his left chest. These are likely Italy’s Silver or Bronze Medals of Military Valor, awarded for frontline combat gallantry. Such decorations reinforce that they saw action in an army combat unit rather than purely support or garrison duty.

Family context: My grandfather received a promotion on 1 Jan 1917 after at 2 deployments in the infantry. Near the end of 1917, he became a prisoner of war for exactly one year. After they freed him from the POW camp, they sent him home. The medal he's wearing should be from his first 2 deployments.

If I had a clear photo of their medals, Copilot said it can identify them.

Location Recognition

Location 1. My aunt's photo collection has photos of a steel mill where my Grandpa Pietro must have worked. After some detective work, I found the most likely location and the name of the company. Copilot can tell me if I'm right.

I asked Copilot, "Could this be the Republic Iron and Steel Works that used to be in Youngstown, Ohio?" and dropped in a photo that shows a mill with 8 smokestacks. Copilot said:

The scene you’ve captured—multiple tall smokestacks rising above a large multi-bay brick mill, with freight tracks running right at the front—matches exactly the classic layout of the Republic Iron & Steel Company’s Youngstown Works.

It told me more about the mill and then said:

…its mills sat just south of the Mahoning River with the railroad cutting through the foreground—exactly as in your photo.

Location 2 and beyond. I have an old family vacation photo. I'm guessing it shows a building in Williamsburg, Virginia, because I know we went there. But I'd like to know for sure. I asked Copilot, "Is this Williamsburg, Virginia?" and dropped in the photo.

Copilot confirmed that it is Colonial Williamsburg and said it's the Governor's Palace. I went there at age 11, and now I know for sure this is it.

As a test, I gave Copilot a few photos I can identify to see what it thought:

  • Disneyland in 1961. Copilot said, "That’s Disneyland’s Sleeping Beauty Castle, centered in Fantasyland. In 1961 it was still brand-new—only eight years after the park opened—its pastel turrets and fluttering pennants shining against the California sun."
  • Rome's Pantheon. Copilot said, "You’re looking at the Pantheon in Rome, Italy."
  • Pemaquid Point Lighthouse, Bristol, Maine. Copilot said, "You’re looking at Pemaquid Point Lighthouse on the rocky shores of Bristol in Lincoln County, Maine."

But I stumped Copilot with 2 locations, and the second one surprised me:

  • Chuang Yen Monastery in Carmel, New York. This isn't a famous monastery, but Copilot felt it was in Henan Province, China, or Kaohsiung, Taiwan.
  • Locust Grove Estate, one-time home of Samuel Morse in Poughkeepsie, New York. Copilot said, "You’re looking at Cedar Grove—the home and studio of Thomas Cole, founder of the Hudson River School of painting. It sits at 218 Main Street in Catskill, on the west bank of the Hudson River." Ooh. Wrong side of the river and wrong style of architecture. The only things the 2 houses have in common are a Hudson River view and a pale yellow color.

Whatever AI tells you about your photos, remember to use your own detective skills and follow up on what it says. Don't forget to think about the photos in a family tree context for more insight.

2 comments:

  1. Is there a similar product for Apple users?

    ReplyDelete
    Replies
    1. It's a website -- https://copilot.microsoft.com/ -- and you can even sign in with your Apple ID.

      Delete