Why identifying AI deepfakes is becoming an impossible task for the human ear
In short: Even experts like Hany Farid, a digital forensics pioneer, struggle to distinguish their own voices from AI-generated clones. Generative artificial intelligence has transformed deepfakes from crude, time-consuming experiments into cheap, scalable tools of deception. As these fakes infiltrate daily life, the speed of misinformation is outpacing our ability to verify truth.
Hany Farid, a professor at the UC Berkeley School of Information and Chief Science Officer at GetReal Security, has spent years analyzing the digital traces left by synthetic media. He argues that the modern deepfake ecosystem has evolved significantly from its origins in nonconsensual sexual imagery. What was once a niche, labor-intensive practice is now a consumer-grade feature capable of producing convincing video-call body doubles or mimicking a child’s voice during a 2 A.M. phone call.
Farid is notably skeptical of the 'AI mystique,' preferring to describe these systems as 'token tumblers' rather than sentient creators. He warns that relying on filtering or detection software is not a complete solution. Because the technology is designed to be fast and scalable, the damage caused by a deepfake is often irreversible by the time the content is officially disproved. The speed of dissemination is the primary weapon in this new landscape of digital deception.
Rather than focusing solely on identifying individual fakes, Farid advocates for a structural approach to the problem. He suggests that if society wants to maintain a world where evidence remains credible, we must shift our focus toward rebuilding rules of liability. By targeting the economic choke points that make digital fraud both cheap and profitable, he believes we can blunt the impact of these systems. Ultimately, the challenge is not just technical but systemic, requiring a fundamental change in how we regulate the infrastructure of digital media.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Summary by Intellect · source: How AI Deepfakes Are Really Made | Hany Farid
- Also: A deepfake can ruin you before breakfast
To cite this card, link to https://intellectplus.co.uk/app/ideas-how-ai-deepfakes-are-really-made-hany-farid.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Artificial intelligence · More in Ideas & Philosophy · Latest · A–Z · All topics
