Twitter to label deepfakes and other deceptive media

If they get the right false positives, false negatives, and bias, it might enhance their media reputation overall. For example, a while ago the White House posted a video of an altercation with a journalist which a lot of journalists believed had been manipulated to make it look worse than it was. Now, this obviously wasn't true on looking at the video, and indeed a couple of outlets hired video forensics experts who concluded this - but AP managed to find a non-expert who was merely a video editor and confirmed their belief by mistaking signs of frame rate conversion for the video being sped up, and the press ran with it.

Meanwhile, there was an a very convincing deceptively-edited video circulating on Twitter which clamed to show parts had been sped up by overlaying the White House version on the original - it was very vivid, with the overlay seeming to jump ahead in fast-moving parts. This was an illusion created by running the videos 7 frames or so out of sync the whole way through and abusing the fact the slight differences weren't noticable in slow-moving parts during normal playback. No-one noticed. Youtube fighter of misinformation Captain Disillusion even helped spread it and gave his approval.

My general reckoning is that Twitter would get much more praise for labelling the non-tampered video as misleadingly edited and endorsing the accuracy of the video that was misleadingly edited to make the other one appear that way, than the other way around.

1. how is this not a slippery slope?

2. have we completely given up on critical thinking and education?

3. free speech vs not so free speech. i guess we're choosing "not so free"?

1. How is allowing the unchecked proliferation of false and misleading media not also a slippery slope, and why is that never a concern?

2. False dichotomy - Twitter attempting to label deepfakes and deceptive media has no negative effect on critical thinking or education, just as Twitter not doing has no positive effect on either. Twitter is not setting education policy for any society or government, here.

3. There is no government or society in the world which does not place some limits on speech in some form, even the US has laws against terrorist threats, slander, libel, lying under oath, misleading labels on food and drugs, misleading contracts, etc ... demonstrating that one common and widely accepted basis for limiting speech is deception and misleading information.

Also, merely classifying data is not preventing its spread, and is not, therefore, a limit on free speech even in theory. If that were the case, then every tag on Twitter would also somehow be "not free speech."

Also, Twitter is a privately owned platform, not a government or a public square.

> Attempting to label these so people aren't manipulated by it is something that deserves praise.

it's pretty incredible how we went from "give people the benefit of the doubt" to "people are manipulated".

>3. free speech vs not so free speech. i guess we're choosing "not so free"?

Well, to be fair, it's all free speech.

The guy who uploaded the deep fake exercised his right to free expression to do so, and the private company that labeled it a deep fake also exercised their right to free expression. Technically speaking, this is an example of the radically open and unregulated nature of free speech in the private sector. Other people labeling your content as "lies" or whatever is a natural consequence of that radically open environment.

It would be "not so free" if the government sent people around to question you whenever you upload a deep fake. Like they do when people mouth off while in, say, a CounterStrike session, about what they're going to do to the other kids at school.