Dartmouth's Hany Farid: Assume Every Video Is Fake Until Proven Real

I Don't Trust Visual Recordings Anymore. But One Day I Will Again

Dartmouth's Hany Farid: Assume Every Video Is Fake Until Proven Real

Deepfake detection pioneer Hany Farid argues the burden of proof has flipped: creators must now prove their footage is genuine. After a viral Tomahawk missile video forced him to spend a Saturday on frame-by-frame forensics, he concluded that forensic analysis alone can't scale to internet speed. His answer is provenance signals—C2PA content credentials, Google's SynthID watermarks, and Apple's cryptographically signed reference images—that let good actors prove authenticity rather than chasing bad ones.

In today's world of generative AI, however, I think we need to shift this stance to a 'presumption of fake.' That is, the burden is now on the creator to prove that their content is real.
  1. fedpost

    These signing schemes are a very long way off right now. Even if they get widely adopted and if they aren't subverted immediately, it's going to take a decade or more for widespread adoption. Most of the "crazy shit" footage you see online is from an old cellphone that's not getting replaced any time soon.

  2. lowbloodsugar

    Do we imagine that the keys from a DSLR or phone can’t be extracted by a sufficiently motivated actor? Or that the sensor on such a device couldn’t be replaced with a substitute image generator?

  3. nayuki

    > Credentials that follow the C2PA standard, for example, attach a cryptographically signed manifest to a photo or video at the moment of capture; this metadata records the device, the time and location of the recording

    This claim sounds like a tidy solution, but my gut feeling is that it doesn't really solve the problem. I struggle to articulate exactly why I think it's wrong, but I'll list some ideas:

    * The camera presumably holds an asymmetric private key. The camera is in the user's hands, and chips could be removed and examined to extract the key with immense difficulty. Extremely few people in this world have the knowledge and equipment to do this, but it only takes one person to break the system for everyone.

    * The camera maker might leak the key through negligence. Look at other examples where companies leaked TLS/SSL private keys or other cryptographic secrets.

    * This scheme seems to give a lot of power to a small number of companies to define image authenticity in a proprietary way. This gatekeeping doesn't seem like a good direction for society.

    * If someone is determined to break the system, eventually someone does. Look at the history of DRM and how countless schemes have been broken. Look at how Intel SGX was broken (it claims to allow a CPU to execute secret computations on behalf of someone else without the physical hardware owner being able to inspect those computations).

    * Some image authentication schemes were cracked, such as ElcomSoft breaking an older ve […]

More from this day

2026-11-21