The world’s top deepfake artist is wrestling with the monster he created
Hao Li has spent his career perfecting digital trickery. Now he’s working to confront the problem of increasingly seamless off-the-shelf deception, reports Will Knight.
Raising awareness: Li has been trying to get CEOs and politicians around the world to consider the consequences that videos doctored with AI—“deepfakes”—could have for them, and for the rest of us. They are virtually guaranteed to be weaponized.
About Li: He isn’t your typical deepfaker. He’s edited some of the most famous faces in the world for modern blockbusters, fooling millions of moviegoers into believing in a smile or a wink that was never actually there, or adding their faces onto stunt people’s.
Breakthrough: In 2009, Li developed a way to capture a person’s face in real time and use it to operate a virtual puppet, without needing motion-trackers.
Troubling potential: Even as Li races ahead with digital fakery, he’s troubled by the potential for harm. He’s working with researchers refining tools to detect deepfakes. But he thinks it’s just a matter of time until deepfakes are totally undetectable.
See the full story here: https://www.technologyreview.com/s/614083/the-worlds-top-deepfake-artist-is-wrestling-with-the-monster-he-created