A couple of years ago in the run up to the 2016 US election, I happened to stumble upon Newt Gingrich being interviewed on some US cable news program. He was stumping for Trump and the interviewer was quizzing him about one of the innumerable falsehoods the Candidate was peddling on the campaign trail.
“It simply isn’t true,” the newsman said.
“That doesn’t matter,” replied Gingrich smugly, “What matters is people believe it’s true.”
This was early in the campaign, before fake news became the mantra of Trump’s MAGA movement, but as the months wore on and events south of the border became increasingly surreal and detached from reality, Gingrich’s comment came back to me again and again, more troubling with every visitation.
Troubling, because it was an affront to everything I hold to be true. Against all odds, against all evidence, against all fact: people will believe what they want to believe.
(Considered from a meta-viewpoint, it occurred to me that this is perhaps the mystery that underpins all religious faith: when there is absolutely no empirical proof of something, yet people believe it anyway. What did I read recently? When one hundred people believe a falsehood, it’s fake news. When one billion believe, it’s a religion. Discuss.)
Bottom line is that we are living in a society that doesn’t know what to believe. But if you thought things were bad now, they’re going to get worse. Check this out.
Let me try to parse this. This is video of Jennifer Lawrence, an actress, answering press questions following a recent awards ceremony. The thing is, her face has been replaced with that of Steve Buscemi, also an actor (and an excellent one) but who is completely irrelevant to the topic at hand.
How was this done? Here is my rather old school explanation.
We have video of Jennifer Lawrence delivering the speech.
We have video of Steve Buscemi speaking.
Computer programs analyse these images and create 3D models of both faces, then, using AI, recreate the motions and expressions of JL’s face model using the data of the SB face model, then seamlessly composite it back onto the original footage.
There are several layers to this process that are frightening to me.
That it can be done at all. This is a reality we all must accept. It is here, it is happening.
That it can be done so quickly and so easily. Ditto. Not so long ago these capabilities were limited to a few highly skilled and exotically equipped visual effects houses; now they are available to almost anyone, anywhere.
What it means for the future of media. This is the big one. In the midst of our fake news addled / conspiracy theory ridden / insanely partisan media environment, deep-fakes add yet another layer of artifice to the increasingly shaky notion of Truth. Heretofore, we could dismiss text based news stories as fabrications, we could disparage still images as photoshopped fakes. But video footage enjoyed a certain bullet-proof veracity. We could believe it.
Until now.