| Agree, and it's even worse than that An incredibly small minority of people even understand your phrase with any actual fidelity and depth of meaning: >>it can be verified there (including cryptographically, say by signing the file or even using HTTPS) Even fewer of that microscopic minority have and understand how to use the tools required to verify the video cryptographically, AND even fewer know how to fully validate that the tools themselves are valie (e.g., not compromised by a bogus cert). Worse yet, even in the good case where everyone is properly skeptical, and 90+% of us figure out that no source is trustworthy, the criminals have won. The goal of disinformation is not to get people to believe your lie (although the few useful idiots who do may be a nice bonus). The goal of disinformation is to get people to give up an even seeking the truth - just give up and say "we can't know who's right or what's real" — that is the opening that authoritarians need to take over governments and end democracies. So yes, this is next-level-screwed material. |
Kind of, but once you have a single verifiable channel back to the source (in this case, some statement by the president) it's now possible for anyone to construct a web of trust that leads back to that source. For example, multiple trustworthy independent outlets reporting on the same statement in the same way, providing some way to locate the original source. This is why new articles that do not link to (on the web) or otherwise unambiguously identify a source wind me up. "Scientists say" is a common one. It's so hard to find the original source from such things.
This falls over in two ways: sources become non-independent and/or non-trustworthy as an ensemble. Then you can't use them as an information proxy. This is what is often claimed about "the mainstream media" and the "non-mainstream media" by the adherents if the other. All the fact checks in the world are worthless if they are immediately written of y those they are aimed at as lies-from-the-system.
The second way is that people simply do not care. It was said, it sounds plausible, and they want to believe it.
So I would say that actually the risks here are social, not technological. Granted, perhaps a deepfake-2'd video might convince more people than a Photoshopped photo. The core issue isn't the quality of the fake, it's that a significant number of people simply wouldn't care if it were fake.
Doesn't mean we're not screwed, just not specifically and proximally because of falsification technology, that's accelerant but not the fuel.