Deepfakes and Misinformation
For most of human history, a photograph or a recording of someone's voice was treated as proof. If you saw a video of a person saying something, you believed they said it. That old rule is now broken.
Generative AI can fabricate faces, voices, photographs and video that pass a casual glance. This is not science fiction and it is not rare — ordinary people are already being scammed, embarrassed and misled by it. The good news: you do not need to become a forensics expert to protect yourself. You need a habit of checking, and this lesson gives you one that works even when the fakes get better.
What you will be able to do after this lesson
- Explain what a deepfake is and tell apart the two families: swapping a real person's face/voice versus a fully synthetic person who never existed.
- Describe, in plain intuition, how these fakes are made — enough to understand the capability, with no recipe for making one.
- Name the main harms — voice-clone scam calls, fake statements, non-consensual images, fabricated "evidence", and the sneaky liar's dividend.
- Explain why our brains fall for it and tell misinformation apart from disinformation.
- Apply the SIFT method to a suspicious photo or video, read laterally like a fact-checker, and run a reverse image search.
- Follow a personal before-you-share checklist and set up a family code word against voice scams.