A dramatic photo lands in your family WhatsApp group with a message saying it shows a flood, a protest or a celebrity in your city today. It looks convincing. But many viral images are old pictures from another place, edited photos or, increasingly, images made with AI. A reverse image search is one of the quickest ways to check. Instead of typing words, you search with the picture itself and see where else it has appeared online.
This guide explains how to do a reverse image search on a phone and a computer, which tools to use, how to read the results, how to check videos and an illustrative verification you can practise with. It is useful for journalists, content creators, social media managers and anyone who wants to share responsibly.
What is a reverse image search?
A reverse image search engine compares your image with the images it has indexed and shows pages that contain the same or similar pictures. That can help you answer three key questions:
- Is this image older than the claim says? If it appeared online years ago, it cannot show today’s event.
- Where did it first appear? The earliest versions often lead to the original photographer, news report or context.
- Has it been changed? Comparing versions can reveal cropping, added text, swapped faces or other edits.
A reverse image search does not prove an image is real. It gives you leads that you then check.
Tools you can use
- Google Lens and “About this image”: on Android or iPhone, open the Google app, tap the Lens icon and choose a photo from your gallery or a screenshot. Google’s help page on searching with an image explains the steps. Google’s About this image feature, available in some regions, can show when Google may have first seen the image, other pages that use it and, when available, details on whether it was made or edited with AI.
- TinEye: TinEye lets you upload an image or paste its link, then sort matches so the oldest appear first, which is helpful for finding the earliest version.
- Bing Visual Search: Bing Visual Search often finds different matches from Google, so it is worth trying both.
- InVID-WeVerify plugin: the InVID verification plugin for Chrome and Firefox is widely used by fact-checkers. It can split a video into keyframes and run reverse searches on several engines.
Different engines index different parts of the web, so use at least two before you draw a conclusion.
How to do a reverse image search: step by step
- Save the best copy you can find. Download the image or take a clean screenshot. Crop out app borders, but keep the main subject.
- Search with the whole image first in Google Lens and one other engine.
- Search with crops. If the picture includes a distinctive building, sign, logo or face, crop to that part and search again. Crops often find matches the full image misses.
- Sort by date where possible. Look for the earliest result and open it. Note the date, the website and the caption.
- Compare versions. Look carefully for differences in text, weather, clothing, signboards or objects added or removed.
- Check the context. Does the earliest source describe the same place and event as the viral claim? Is it a reliable outlet?
- Look for existing fact-checks. Search the claim with words such as “fact check”. In India, PIB Fact Check publishes checks on claims about the central government.
- Decide and record. Write down what you found, with links, before you share or publish anything.
How to check a video
Videos cannot be searched directly in most engines, but you can search their frames:
- Take screenshots of several distinctive moments, especially the opening frames and any shots with signs or landmarks.
- Use the InVID keyframes tool to extract frames automatically and search them.
- Listen carefully for language, accents, announcements or music that may not match the claimed location.
- Check the uploader: when did the account first post it, and does it usually share original footage?
Reading the results: common patterns
- Old image, new claim: the same photo appears in a news report from years ago. This is one of the most common forms of misinformation.
- Right image, wrong place: a real photo from another country or state is shared as local news.
- Edited image: the earliest version looks different, for example with a different banner, crowd size or face.
- No matches at all: this does not prove the image is original. It may be new, rarely shared or AI-generated. Look for other signs and other sources.
Spotting possible AI-generated images
AI images keep improving, so visual clues are not always reliable. Still, look for distorted hands or text, unnatural reflections, repeated patterns, inconsistent shadows and backgrounds that melt into each other. Check whether the image has any provenance details, such as those shown in Google’s About this image panel, but remember that metadata can be removed or changed. Many social platforms also strip original photo data when images are uploaded. If you create AI images yourself, our guide on how to write AI image prompts explains why clear labelling matters.
Illustrative example: checking a viral photo
This example is illustrative and invented. A photo is shared with the caption: “Heavy flooding at the main bus station in our city this morning.”
- Step 1: you save the photo and search it in Google Lens. Results show similar images on several news sites.
- Step 2: you search a crop of a signboard in the picture on TinEye and sort by oldest. The earliest match is a news report from four years ago, about a different city.
- Step 3: comparing the two, the viral version has been cropped to remove the other city’s name on a shop sign.
- Step 4: you search for the claim with “fact check” and find no report of flooding at your bus station today; local news shows normal traffic.
- Conclusion: the image is real but old and from another place. You reply in the group with the link to the original report, politely and without blaming the sender.
Reverse image search checklist
- Did I search with at least two engines?
- Did I try crops of distinctive details?
- Did I find and open the earliest version?
- Does the original context match the claim?
- Have I looked for existing fact-checks?
- Have I recorded my sources before sharing?
Where these skills lead
Verification is a core skill in journalism and digital media. If you enjoy this kind of work, read our guides on how to become a fact-checker and how to become a photojournalist. For structured training in media and digital skills, explore our programmes and apply here. If you would like to support women building media careers, you can also donate to Karuna Media.
The Giving Movement by Karuna Media · #TheGivingMovement
Wear the cause: tees, a tote and a mug that fund media training for women
Tee (Black) ₹699 · Tee (White) ₹649 · Tote ₹449 · Mug ₹399. Printed on demand and shipped across India. 100% of the profit funds media training for underprivileged women.
Buy now on Razorpay Prefer to donate? Go to our donate pageMerch is a purchase, not a donation, and is not eligible for 80G. To make a donation, use our donate page.
