A Misstep. Zooms. The AI Keeps Coming.
A man makes a mistake. He accepts responsibility and subjects himself to a series of Zoom-a-culpas. People in his audience tell him that, “We hate AI!” (Thought bubble over man’s head: “No s**t.”) He’s lost the room. Facts established beyond dispute are now disputed. He’s being called to account for developments utterly beyond his control. Hints of narrator boycott are in the air and on social media. An admission from the Zoomer of Ceremonies: “We’ve only used our T2S algorithm on two paragraphs—and it didn’t work well.”
Indeed, this whole thing isn’t working well.
[Disclosure: I wasn’t there. This is all secondhand buzz, if you know what I mean . . . ]
Q: Has any of this altered the current state of affairs? No.
To wit: AI is here. It’s not going away. And it doesn’t care whether you like it or not. AI doesn’t care about your art, your living, or you. It doesn’t care that you quit this organization or refuse to work for that business. (And it sure as hell doesn’t care about Eljin Productions and John McElroy.)
AI just is, and software engineers will continue to develop it while we grind our teeth to nubs in rage. So anger is pointless.
My hope: We take things down a notch.
To that end, a few notes:
T2S algorithms aren’t ready to take over the audiobook industry any time soon. But it’s a good bet that, as far as our industry is concerned, the movers in AI will be big players who may or may not have anything to do with the industry right now. Such companies have the capital to buy small, sophisticated tech firms fully-dedicated to T2S. And when a Big One moves on this, trust me: There will be no Zoom-a-culpas, no gestures to keep vendors on board.
Implementing AI would demand a significant reorganization of production and distribution. I was tooling around with this idea in an earlier post. I don’t have special insight here, but I’m confident about this: It would require serious effort and lots of resources to manage the politics, finances, practical mechanics, and possible legal challenges of such a shift. It would take a lot of time and pain. (BTW: Major audio publishers are adding staff to their existing operations—a signal of continuity not change.)
So . . . a percentage of the couple of million books published in the US yearly is offered through AI. Why does that necessarily displace anyone now working? More audiobooks is likely a good thing for anyone on the production side. The economics of a virtuous cycle: AI reduces production costs and encourages publishers to release titles that would never, ever be produced using a narrator. Too expensive, no return. This would attract new listeners with new interests, expand demand for audiobooks on the whole, generate new cash flows that could be directed to studio-produced programs once considered too risky to publish. Why? Because those programs, particularly fiction, will be more entertaining. The result: more revenue and more narration jobs than ever. (A little Pollyanna-ish, perhaps, but no more outlandish than presuming a vicious cycle. Yeeeeeek! Robo apocalypse!)
Mr. Automation himself, Elon Musk, thinks that a real-live narrator, Grover Gardner, rocks! Apparently, performance matters even to an uber Master of Scale.
If and when T2S becomes marketable in the audiobook space, personalized digital replicas may offer performers a new way to license their work. The legal framework exists and is enforceable. The labor contracts could be worked out. All that’s needed is business logic to make the payments possible—that and functioning algorithmic products and marketplaces!
SAG-AFTRA and the publishers are protecting performers’ rights in the AI realm. The side letter to the new agreement demonstrates that. People can’t steal your voice and make money out of it. You (and your heirs) have the right of publicity. It’s in the law and the agreement.
Something worth remembering: Little more than a decade ago, studio-based performers in NY and LA fulminated against those barbarians in their home studios. Uncouth. Non-union. Un-actorly. Cheap. Today, home-studio narrators (many of them union) account for the great majority of audiobook performers. They transformed the work. They disrupted the markets for talent and studios—and producers like me. They adopted (and adapted to) emerging technologies and enabled much of the industry’s subsequent expansion. They organized and put a floor under rates. Today, they’re highly-regarded and sought-after. Once the avant-garde, they’re now the establishment. The point: Change happens. We’ve gone through it before. We’ll see it again. You can’t stop change. A cliché that’s nonetheless true.
Look: I’m as unsettled as anyone. But standing on a beach yelling at the oncoming tsunami won’t stop the inevitable. And no one has a crystal ball. T2S may involve more creation than destruction, more opportunities than fewer. No one knows how this will play out.
Cool heads. Plenty of time to adapt and prepare. Let’s see where this goes.
And Happy Thanksgiving! (That wish is real not artificial.)
