Reverse Image Search
MinutesUpload or paste an image (or a video frame) into an image search engine to find where else and when it has appeared, which exposes recycled photos, stolen profile pictures, and images given a false caption.
How to do it
- 1Save the image or copy its address. On a phone, a long press usually offers "search image"; on a computer, right-click. For video, pause on a distinctive frame and take a screenshot.
- 2Run it through more than one engine (Google Lens, TinEye, Bing, Yandex). Their indexes differ, and they differ in strength on faces, places, and older material.
- 3Look for the oldest appearance. TinEye lets you sort by date; elsewhere, add an earlier date range to the search. An image older than the event it is said to show settles the question.
- 4Compare captions across the copies. If the same photo is labelled as three different cities, none of the labels can be trusted without the original.
- 5Crop and retry. Searching a cropped region (a sign, a building, a uniform patch) often finds the source when the full frame does not; mirrored copies are common, so try flipping it too.
- 6For a profile picture, a match to a stock-photo site, a model's portfolio, or a different name is strong evidence that the account is not who it claims to be.
What to say
- “Before we share that photo, give me a minute to see where else it has been.”
- “That picture of you is lovely. I ran it through a search and it also appears under another name; can you explain that?”
When to use it
- •Dramatic photos or clips during disasters, protests, and wars.
- •A new online romantic interest, seller, landlord, or recruiter whose photos look professional.
- •Product and rental listings, charity appeals, and "before and after" images.
- •A screenshot or meme that attributes a scene to a specific time and place.
Counters
Evidence and how strong it is
Reverse image search is standard practice in newsroom verification (Silverman 2014) and in the fact-checking routines taught by Caulfield & Wineburg (2023); consumer-protection agencies recommend it for checking the photos of online romantic contacts. Research supports the premise that the threat is real: Hameleers et al. (2020) found that pairing false text with an image made disinformation slightly more credible than text alone, and Wardle & Derakhshan (2017) identify genuine images in a false context as a leading type of misinformation. Evidence strength: practitioner consensus. There are many documented cases in which the method identified recycled or stolen images, but no controlled trial measures how much it improves ordinary users' accuracy, and its hit rate is unknown.
- No match proves nothing. Private images, new images, and images generated by AI have no earlier copies to find; an absence of results is not authentication.
- A match shows the image existed before; it does not show which caption is right. Trace to the earliest, most specific source (the photographer, the agency, the original post).
- Face-matching services that identify strangers raise serious privacy and safety concerns and are error-prone. Use image search to check claims made to you, not to unmask private individuals.
- If a search shows that someone you are emotionally or financially involved with is using stolen photos, stop sending money and tell someone you trust. If you are being threatened with intimate images, do not pay; preserve the messages and report to the platform and to police or your national reporting centre.
- Silverman, C. (Ed.) (2014). Verification Handbook: A Definitive Guide to Verifying Digital Content for Emergency Coverage. European Journalism CentreNewsroom procedures for verifying images, including reverse image search to establish provenance and date.
- Caulfield, M., & Wineburg, S. (2023). Verified: How to Think Straight, Get Duped Less, and Make Better Decisions about What to Believe Online. University of Chicago PressInstruction in tracing images to their original context as part of everyday web verification.
- Hameleers, M., Powell, T. E., Van Der Meer, T. G. L. A., & Bos, L. (2020). A Picture Paints a Thousand Lies? The Effects and Mechanisms of Multimodal Disinformation and Rebuttals Disseminated via Social Media. Communication Research, 47(2), 281-301Experimental evidence that adding images to false text modestly increases its credibility.
- Wardle, C., & Derakhshan, H. (2017). Information Disorder: Toward an Interdisciplinary Framework for Research and Policy Making. Council of Europe report DGI(2017)09The classification of genuine imagery shared in a false context as a major form of misinformation.