Misinformation spreads fast, and manipulated images are one of its favourite vehicles. Whether you are a journalist fact-checking a breaking-news photo, an online shopper verifying a product listing, or simply curious about a viral image—knowing how to tell a real photo from a fake one is an essential digital skill.
This guide walks you through a 5-step verification workflow you can complete in under five minutes using free online tools. No software installation required.
Why photo verification matters
AI image generators like Midjourney, DALL·E, and Stable Diffusion produce photorealistic output that fools most viewers at a glance. Meanwhile, classic editing techniques—splicing, cloning, and retouching—remain just as common. The result: you can no longer trust a photo at face value.
A structured verification workflow reduces guesswork and gives you a defensible conclusion, whether you are publishing a story or deciding to share a post.
Step 1 — Check the EXIF metadata
Every photo created by a real camera embeds metadata: the camera model, lens, shutter speed, date, and often GPS coordinates. AI-generated images and heavily edited photos typically lack this data or contain inconsistencies.
Open the image in the PhotoRadar EXIF Viewer. Look for:
- Camera make & model — Missing? The image may be a screenshot, AI render, or stripped of metadata.
- Software tag — Values like "Adobe Photoshop" or "GIMP" indicate post-processing.
- Date/time — Does it match the claimed event? Check time zones too.
- GPS coordinates — If present, verify they match the claimed location on a map.
Missing metadata alone does not prove a photo is fake—social media platforms strip EXIF on upload. But it removes one layer of trust and makes the next steps more important.
Step 2 — Run a Reverse Image Search
A Reverse Image Search checks whether the image already exists elsewhere on the web. This catches:
- Recycled images — An old photo repurposed for a new claim.
- Stock photos — Sometimes passed off as original journalism or evidence.
- Earlier versions — Finding a higher-resolution or uncropped original can reveal edits.
Upload the image and review the matched pages. Pay attention to the earliest publication date and the context in which the image originally appeared.
Step 3 — Scan with an AI Image Detector
The AI Image Detector analyses pixel-level patterns that generative models leave behind. These include:
- Unnatural skin smoothness or texture repetition
- Warped edges around hair, glasses, or teeth
- Inconsistent lighting and shadow directions
- Subtle grid artefacts from diffusion models
The tool returns a confidence score. A high "AI-generated" score is a strong signal, but always combine it with the other steps—no single detector is infallible.
Step 4 — Inspect the Tamper Heatmap
Even if a photo started as a real capture, it may have been doctored afterwards. The Tamper Heatmap analyses JPEG compression artefacts across the image. Regions that were added, removed, or altered after the initial save show different compression signatures.
Look for brightly highlighted areas—especially around people, text, or objects central to the image's claim. Uniform heatmaps with no hotspots suggest the image was saved in a single pass without local edits.
Step 5 — Cross-reference and conclude
Compile your findings into a simple verdict matrix:
- All checks pass — The photo is very likely authentic.
- One check flags an issue — Investigate further; there may be an innocent explanation (e.g., metadata stripped by a messaging app).
- Multiple checks flag issues — Treat the image as unverified and do not share it as fact.
Document your steps if you are publishing or reporting. A transparent methodology strengthens your credibility.
Quick-reference checklist
| Step | Tool | What to look for |
|---|---|---|
| 1. Metadata | EXIF Viewer | Camera model, date, GPS, software tag |
| 2. Origin | Reverse Image Search | Earliest source, context mismatch |
| 3. AI check | AI Image Detector | Generative artefacts, confidence score |
| 4. Tampering | Tamper Heatmap | Compression anomalies, spliced regions |
| 5. Verdict | Your judgement | Combine all signals |
Frequently asked questions
Can I verify a photo for free?
Yes. PhotoRadar offers free tools including an AI Image Detector, EXIF Viewer, Tamper Heatmap, and Reverse Image Search. All run in your browser with no signup required for basic checks.
How accurate is AI-based fake photo detection?
Modern AI detectors can identify common generative artefacts with high confidence. However, no single tool is 100 % accurate. Combining multiple checks—metadata, reverse search, AI detection, and tamper analysis—gives the most reliable verdict.
Does a photo without EXIF data mean it is fake?
Not necessarily. Social media platforms strip EXIF data on upload, and users may remove it for privacy. Missing metadata is one signal, but you should always cross-reference with other verification steps.