Learn
Learn About AI & Images
Guides covering how AI image detection works, the impact of deepfakes, and how to identify AI-generated photos.
No Label Doesn't Mean Real: What AI and Retouching Labels Cover in 2026
This month's viral fakes of a real woman carried invisible watermarks nobody scrolling past could see. What the EU AI Act, Meta, TikTok and Instagram now require, what France and Norway demand of retouched ads, and why a missing label tells you almost nothing.
Read →
October 26, 2026
Fake Recruiters With AI Faces: How to Check Who Is Offering You a Job
Employment scam losses reported to the FBI went from $70 million to $363 million in two years, and the opening message usually comes with a friendly headshot. Why that face is so often generated, why a reverse image search misses it, and what to check before you reply.
Read →
October 12, 2026
Google's SynthID Detector Is Open to Everyone. What a Negative Result Means
Google has watermarked 180 billion images and will now check a file for you. The badge is strong evidence when it appears, and almost worthless when it does not. We look at the coverage gap, the robustness claims, and the edited-but-real photos that get flagged.
Read →
October 1, 2026
Apple Reference Image: What a Signed Photo Proves
The iPhone 18 Pro signs pixels at the sensor, before any software touches them. We look at how it works, why Apple went around C2PA, and the gap between proving capture and proving truth.
Read →
September 16, 2026
How to Verify a Photo Before You Share It: A Practical Guide
Reverse image search, the earliest copy, Content Credentials, and what detail-level inspection still catches in 2026. An ordered two-minute routine for anyone, with the limits of each step spelled out.
Read →
August 18, 2026
Training Data and Consent: The Hidden Ethical Issue in AI Image Generation
The people whose photos trained today's image generators were never asked. What is actually inside LAION, why copyright and biometric law both miss the subject of a photograph, and what opt-out really does.
Read →
July 21, 2026
The Ethics of AI-Generated People: Who Owns a Face That Never Existed?
A face that never existed has no rights holder — and that accountability gap enables fraud, synthetic diversity washing, and manipulation at scale. We examine the legal, philosophical, and practical stakes.
Read →
July 7, 2026
Japan's 2024 Deepfake Laws: What Changed and What Gaps Remain
In 2024 Japan passed a platform takedown law and issued AI business guidelines, but neither names deepfakes directly. We examine what actually changed and where the legal gaps remain.
Read →
June 16, 2026
AI-Generated Images in Journalism: A Global Crisis of Trust
How AI-generated fakes are eroding trust in news — from the Southport riots photograph to the Ireland election deepfake, how wire services are responding, and what the liar's dividend means for journalism.
Read →
June 2, 2026
Frequency Analysis: How Researchers Spot AI Images at the Pixel Level
How Fourier transforms expose GAN checkerboard artifacts, what traces diffusion models leave in the frequency domain, why JPEG erases the signals, and where frequency fingerprinting is heading.
Read →
May 26, 2026
The Arms Race: Why AI Image Detection Gets Harder Every Year
Why detectors that score 99% in labs drop to 62% on real-world content, what the generalization gap means, how adversarial attacks exploit it, and what a realistic defense looks like.
Read →
May 19, 2026
What Is C2PA? How the Industry Is Fighting Fake Images with Metadata
How C2PA embeds cryptographic certificates inside image files to prove provenance, who is behind it, and why platform metadata stripping remains the biggest obstacle.
Read →
May 12, 2026
Japan's Portrait Rights (肖像権) in the Age of Generative AI
How Japan's portrait rights, publicity rights, and criminal law apply to AI-generated images — and where the legal gaps still leave victims without protection.
Read →
May 5, 2026
How AI Image Generators Work: Diffusion Models Explained
How diffusion models turn random noise into photorealistic images — the denoising process, text conditioning, latent space, and how Stable Diffusion, Midjourney, DALL-E, and Flux differ.
Read →
April 28, 2026
Why Fake AI Photos Are Flooding Social Media — and What You Can Do
How AI-generated fake images now spread faster than corrections, what platforms are doing about it, and practical steps to check before you share.
Read →
April 21, 2026
How to Tell If a Photo Is AI-Generated: 7 Signs to Look For
Research-backed guide to the visual cues that still reliably reveal AI-generated photos in 2025–2026 — and the tells that no longer work.
Read →
April 14, 2026
AI-Generated Profile Photos in Romance Scams: How to Protect Yourself
How scammers use AI-generated faces to build fake online relationships, the red flags that still work, and how to verify a profile before you trust it.
Read →
April 7, 2026
Deepfakes Explained: What They Are and Why They Matter
What deepfakes are, how they are created, and why detecting them has become an important issue for individuals, media, and society.
Read →
March 24, 2026
How AI Image Detection Works
A deep-dive into the signals and techniques modern AI uses to distinguish real photographs from AI-generated or edited images.
Read →
Want to scan a photo and check if it looks authentic?
Try Maat Scan