Bob Odenkirk on the Writers Guild picket line. Credit…Dana Calvo
Dear Commons Community,
Maureen Dowd has a column this morning commenting on the major issue regarding the Hollywood writers strike: artificial intelligence. Entitled, “Watch Out for the Fake Tom Cruise,” she directs our attention to the dangers A.I. poses for many of us. Below is an excerpt.
Read and heed!
“Be careful what you wish for, Hollywood studios, as you mess with the primal force of A.I.
Tinseltown is going dark, as the actors join the writers on the picket line. Hollywood’s century-old business model was upended by Covid and also by streaming, which swept in like an occupying army. Then streaming hit a ceiling, and Netflix and Co. scrambled to pivot.
With a dramatically different economic model shaped by transformative technologies — A.I. is a key issue in the strike — the writers and actors want a new deal. And they deserve it.
The Times’s Brooks Barnes describes the mood of the town as très French Revolution, with writers and actors seething in fury over the Marie Antoinette antics of C.E.O.s and studio chiefs collecting humongous paychecks, frolicking in Cannes and jetting to Sun Valley.
Besides pay fairness, writers want to make sure that they’re not rendered irrelevant by algorithms, and actors want to prevent their digital likenesses from coming under new ownership.
It’s a complex issue. Even as writers are demanding that studios not replace them with A.I., some studio execs are no doubt wondering if the writers are being hypocritical: Will they start using A.I. to help them finish their scripts on deadline?
Chatbots are so proficient — and growing more so every second — that many studio suits are probably itching to bypass the middleman screenwriter.
As Puck’s Matthew Belloni said on MSNBC’s “Morning Joe,” “You can say, you know, ‘Here’s the “Social Network” script. Write me a script, but make it about Elon Musk, not Mark Zuckerberg.’”
Jaron Lanier, the father of virtual reality, has long warned that we were cruising for a bruising. As he told me nine years ago, the lords of the cloud were acting as if they had been inventing a digital brain when what they were really doing was making a mash-up of real brains.
He said that when machines translated one language into another, they were leeching from human translators, taking matching phrases from aggregated data; those translators should have the right to negotiate for compensation for unwittingly feeding the A.I. brain.
He also has made the point that Facebook and other social media companies have been extracting our precious data for years, without giving us payment or any of the other rights a first-class citizen would normally have. He said it would be unfair if Hollywood studios created fake versions of actors and then didn’t pay them.
The compensation issue is now center stage. Sarah Silverman joined class-action lawsuits against OpenAI and Meta accusing them of copyright infringement, saying that they “ingested” her work to train their A.I.s.
The ingesting and synthesizing of words, images and music is going on in giant gulps. Indeed, the day is fast approaching when the digerati will be able to make a whole fake movie.
As Lanier said, “They might say, ‘Make me a movie that’s similar to Tom Cruise’s “Mission: Impossible.” However, make sure that none of the synthetic actors can be mistaken for known actors and make sure that we’re not going to get sued, but let’s go right up to the line.’ That’s not quite feasible today, but I don’t see any reason why it wouldn’t be. It’s just math. And we can do it.”
He said that the Hollywood strikers are just the tip of the iceberg. “People say, ‘Why should we help these fancy, lefty, very well-paid actors? Screw them.’ But if you’re making a living driving a vehicle or working in a place where you use heavy machines like an auto body shop, all kinds of jobs, this is going to create the legal precedents that could protect you in the future, too.” Almost nobody is immune to the risk that A.I. could devalue their economic position, even though A.I. will also have widespread benefits.
“Tech companies would be helped by bringing the whole society into the process of improving how models perform using economic incentives,” Lanier said. But, he added, if we get it wrong on “data dignity,” society will “turn into a misery fast enough.”
“This is really for everybody,” he said of the effort not to be swallowed by A.I. “It might not seem like it, but it really, really is.”