Reality Checker
AI Verification Guides for Images, Videos, Audio, and Synthetic Content
Updated daily. Built for professionals, journalists, creators.
Latest Dispatch
View allThe new wave of AI identity scams: how a single photo becomes a full impersonation
Deepfake video is evolving: the most common tells are changing
When TikTok goes viral with an AI edit: why so many people believe it
AI image detector: what it can tell you, and what it cannot
Deep Dives & Investigations
See allThe new wave of AI identity scams: how a single photo becomes a full impersonation
AI identity fraud is no longer just deepfake video. In real cases, attackers start with a single face image, generate variants, clone a voice, and deploy scripted messages at scale.
Deepfake video is evolving: the most common tells are changing
Many deepfakes now look 'clean'. The strongest checks are contextual: where it came from, how it traveled, and whether alternate sources exist.
When TikTok goes viral with an AI edit: why so many people believe it
Viral content does not need to be perfect. It needs to feel plausible long enough to spread. We break down the exact signals that should trigger a verification reflex.
AI image detector: what it can tell you, and what it cannot
Automated detection is a first pass, not a verdict. We explain what detectors measure, how errors happen, and how to escalate to deeper analysis.
Most Read
- 1Why AI detectors are failing on LinkedIn headshots
- 2The 3-second voice clone scam explained
- 3Identity theft in the age of generative video
Verification Toolbox
Reality Briefing
Get the weekly digest of new fraud vectors and verification techniques.
From the field • Signal Notes
New "Flux.1" model variants seen bypassing standard metadata strippers on Twitter.
Political deepfakes in EU elections show 40% increase in audio-only attacks vs video.
Scammers shifting from "Emergency Family" calls to "HR verification" calls targeting remote workers.