A fourteen-year-old gets a message. Someone took her school photo and made it explicit with a free app. Pay up, or it goes everywhere. The image is fake. The fear is not. And here's the part that should stop every parent cold. She probably won't tell anyone. The threat isn't the picture. It's the silence after it. That silence is exactly what the extortionist is counting on.
According to a UNICEF study across eleven countries, more than one million children had their images manipulated into explicit deepfakes in the past year. In some regions, that's one in twenty-five kids. Three children in an average classroom.
A deepfake doesn't have to fool anyone forever. It just has to scare your kid quiet long enough to win. So the most protective thing you can say tonight is simple. Come to me, and I won't make it worse.
The fix here is a conversation. But what happens when the system meant to protect you can't agree on its own rules?
You tap your phone at a pharmacy. Your I.D. appears. No plastic, no fumbling. That's Europe's new digital wallet, coming to every member state by 2026. But each tap might send your facial photo through systems you never approved. To buy wine. To order a book. Every transaction. Most people using it won't know that's happening.
According to Biometric Update, Spain's privacy regulator ruled your face can't be the only way to verify who you are. The European Commission proposed building a mandatory facial photo into every wallet. Two authorities, opposite answers.
Spain said your face is sensitive. Treat it that way. The Commission said every wallet needs it built in. Nobody's reconciled that, and the clock is running on a few hundred million phones.
When AI touches your safety, who has to prove it actually works? Europe just answered that for your car.
You're driving a wet road on a Friday night. Your lane-assist nudges the wheel back. Under Europe's new AI law, that nudge sits in the same category as a surgical robot. Same testing. Same paper trail. The reason is simple. If the system fails, you could get hurt. So manufacturers have to show their work.
According to WilmerHale's analysis, high-risk AI triggers seven requirements, including proof the training data was tested in rain, at dusk, on faded roads. Not just sunny highways. That proof must exist on paper a regulator can read.
And high-risk doesn't mean dangerous. It means someone had to prove it wasn't. It's a building permit, not a warning label. The skyscraper needs more checks than the shed.
Three stories, one question. Your face, your kid's photo, your car's code. AI now decides things that touch your body and your safety. The fight in all three is identical. Not does it work. But can someone prove it works, for everyone, before it reaches you.
Links to every story and today's podcast deep-dives are in the description. See you next time.
