フェイクニュース And Deep Fakes

The January 2019 cover of the Japanese version of HBR caught my attention.

The English catch-copy shows the words “FACT NEWS” being peeled back to reveal the word “FAKE NEWS”.

One of the key articles in the package is titled “Business In The Age of Computational Propaganda And Deep Fakes” and explains how “deep fakes” are creating new issues around the concept of facticity.

What are deep fakes?

The term combines “deep learning” and “fake” to describe AI-driven human image generation and synthesis.

To illustrate the point, the article includes a “deep fake” example of comedian Jordan Peele ventriloquizing president Barack Obama in a fictional PSA in which the ex-president appears to close by saying “Stay woke, bitches”.

Deep fakes are not illegal, but in some cases, the makers are running up against harassment claims and other civil legal disincentives.

While scientists still haven’t figured out to identify deep fakes with 100% accuracy, there are ways you can avoid falling for them.

For example, if you see a provocative video you should ask yourself, Where did it come from? Has the message been substantiated or corroborated? Does it look real? (The current generation have visible distortions and blurring.)

I don’t know how to say “stay woke” in Japanese.

So instead I’ll just say … お気をつけください !