Member-only story
On a Hyperrealistic AI Avatar (Reciting Hamlet’s Monologue), the Uncanny Valley, and the Fairy-Tale ‘Vale of Tears’
Inspired by Melissa Heikkilä’s article “An AI Startup Made a Hyperrealistic Deepfake of Me That’s So Good It’s Scary” in the MIT Technology Review
Melissa Heikkilä may have been unnerved and even shocked seeing her own hyperrealistic avatar created by Synthesia, a relatively new AI startup, and I was dispirited only after the first two paragraphs of Heikkilä’s article in the MIT Technology Review, but there’s no need to truly despair — or is it? I watched a video of Heikkilä’s avatar reciting Hamlet’s monologue and understood next to nothing of it, even though I was once familiar with it in its entirety and am usually capable of understanding enough of Shakespeare when recited well or performed in a theater play.
But synthetic Melissa is not the problem — only a pointer to real difficulties awaiting us ahead.
To describe a little the first video in Heikkilä’s article, the first thing that struck me is that her avatar’s recitation contained no acting — just a voice going randomly up and down and in various other directions, which was very distracting. Also, there was no gesturing to add to the words…
