NewsBite analysis · AI, Business, Security
ZDNet
A low-tech solution from the past may be your best defense against AI deepfakes
Recent incidents demonstrate the growing sophistication of AI deepfakes, as seen in a case where a company lost $25 million due to fraudulent wire transfers. Traditional methods of verification are no longer sufficient, necessitating a shift toward using physical security keys and verbal passphrases. Experts suggest integrating these low-tech measures with existing security protocols to counter new-age threats effectively.
MonitorConfidence: HighImmediate horizonCybersecurity risk: High
Published 21 Aug 2026, 12:00 UTC7 min read
Visit original article ↗Lead image — ZDNet
Summary
TL;DR
- Deepfakes are becoming harder to detect, leading to large financial losses for companies.
- Companies are encouraged to revert to low-tech solutions for identity verification.
- Verbal passphrases can significantly enhance security against impersonation.
- Experts advocate for using hardware security keys alongside verbal authentication.
- There is a critical need for organizations to adapt their security frameworks accordingly.
Interesting facts
- AI deepfakes can now replicate human behavior with alarming accuracy, making traditional checks ineffective.
- Listeners' ability to spot deepfakes has deteriorated to near chance levels.