1. Your CFO joins a video call and urgently asks you to wire funds. Face and voice look right. What do you do?
Answer: Confirm through a separate channel you already trust. Real-time deepfakes can match a face and a voice. In 2024 an Arup employee in Hong Kong wired about $25M after a call where every colleague on screen was fake.
2. True or myth: deepfakes give themselves away because they don’t blink normally.
Answer: Myth. Early deepfakes had odd blinking. Modern tools fixed most visual tells, so no single visual clue is reliable.
3. A video of a famous entrepreneur promotes a crypto platform with “guaranteed returns.” Most likely:
Answer: A deepfake ad. Deepfake celebrity ads are a common hook for investment scams. Appearing on a major platform doesn’t verify anything.
4. What’s the best way to check whether a viral clip is real?
Answer: Find the original source and independent reporting. Visual inspection is unreliable. Provenance — who published it first, and whether credible outlets confirm it — tells you far more.
5. What are Content Credentials (C2PA)?
Answer: Signed information about where content came from and how it was edited. C2PA attaches verifiable provenance. Its absence doesn’t prove something is fake, but its presence helps confirm the origin.
6. Which is harder for a scammer to fake today?
Answer: A question only the real person can answer. Faces and voices can be synthesised from public clips. Shared private knowledge, or a pre-agreed code word, is much harder to fake.
7. True or myth: if a video call is in sharp HD with good lighting, it’s real.
Answer: Myth. Quality isn’t evidence. Generators produce high-resolution video, and some deliberately add “bad connection” artefacts.
8. On a suspicious video call, the most reliable step is to:
Answer: End the call and call back on a number you already know. Real-time tools can say anything you ask. A call-back through a channel the attacker doesn’t control breaks the attack.
9. Deepfake detection tools are best used as:
Answer: One signal among several. Detectors make mistakes in both directions. Combine them with context, provenance and verification — a weight-of-evidence approach.
10. Who do deepfake fraudsters most often target at a company?
Answer: Staff who can approve or send payments. Attackers impersonate executives in order to pressure the people who can actually move money.