How can you tell if a photo or video is real?
Misleading images abound online!
But with careful observation, lateral reading, reverse image searching, and the occasional assist from Google Street View, you can discover the truth behind almost any online image.
In the sections below, you will learn some sophisticated tips and tricks for photo and video verification.
But the most powerful tools you have to defend yourself against visual misinformation are observation, common sense, and critical thinking.
This short slide show to the right introduces you to the basics of thinking through a meme or viral image.
The first step is always to slow down and look!
 |
 |
 |
 |
Journalist Laura Garcia shows uses her powers of observation, reverse image searching, Google Street View, and a few simple searches to figure out exactly where the video really came from.
This video from the Poynter Institute shows the use of more sophisticated tools, but they're all available free online, and you can use them yourself with a little practice.
In this short video, Ms. Goldberg shows you a few ways to do a reverse image search to find the original context for an image, debunk a fake photo, or use it to identify objects in a picture you took yourself.
This video from John Silva of The News Literacy Project shows you how to use Google Street View to geolocate an image you want to know more about and debunk fakes. (double-click to see the whole screen).
Use the images on this page to practice with Google Reverse Image Search, TinEye, or Yandex Images. Where were these images taken? When? What have others said about them?
The Library of Congress has millions of photos in its collection, but it does not know who all the people in the photos are. They have asked the public to help. So far, people on the Internet have identified several hundred! They're down to the last 25. Can you win the Mystery Photo Contest?
First Draft quizzes are TOUGH! You'll need all your powers of observation, plus geolocation and search skills, to find out where and when these photos were taken. Try Verification first, then, for more challenge, Observation and Geolocation.
Bellingcat and journalists around the world post daily verification challenges to sharpen their skills in observation, geolocation, and chronolocation. Think of these as a pro workout!
You've used Reverse Image Search. You've used your observation and geolocation skills to try to determine where the photo was taken. You've Googled the people or places in the photo to try to get the real story behind a photo. But if you're still not sure, here are some ways to examine online images for signs they have been manipulated.
The Fact Checker team at the Washington Post explains how real videos can be decontextualized, edited, or manipulated to create false impressions, and how deepfake videos can manufacture events that never happened.
This video from PBS's science show, NOVA, explains how artificial intelligence has made deepfake videos easier and cheaper to create. Experts are worried about the effect of these videos on political stability.
But help is on the way! InVID Verification is a sophisticated new tool available as a Chrome plug-in that allows users to analyze online video to detect manipulation, geolocate photos and videos, and perform several kinds of analyses. Scroll down to see videos about how it works and to learn how to use all the features.
Watch Frame By Frame, as the name implies, allows you to watch any YouTube or Vimeo video frame by frame or in slow motion. Why? So you can use the same observational skills you would use for photos to detect moments when the video might have been manipulated. You can also reverse image search a particular frame to help you identify place and time.
The Deepfake Lab is a project created by communications designers and researchers to show the public how deepfakes work and how to spot them. These ten very short videos demonstrate how deepfakes are made and invite you to make your own. Click on the "Technical Details" and "Visual Flaws" tabs to see explanations, and compare the original videos to the deepfake versions.