Littwin: Now that the AI threat has pushed Hollywood actors to go on strike, it may be time to worry

On your list of things to worry about — for me, as of this moment, there’s that flight I’m taking on United next week, Russell Wilson’s feud with Future the rapper and, OK, maybe global warming — the dangers of artificial intelligence might not have been keeping you up at night.

But that was before.

Now it’s serious. 

Now that the actors have joined the writers in their two-month-old strike against Hollywood producers, basically shutting down the entire industry, you’re completely engaged. And why not?

This means it could be years before we get to see “Mission: Impossible — Dead Reckoning Part Two.” Are you prepared for that? I don’t know about you, but I would get shaky when I had to wait a whole week to see the next episode of “Succession.” We live in the days where binging has nothing to do with drinking.

But no one will be surprised if the Hollywood strikes last approximately forever. And the AI issue is at the very heart of the problem for both unions. 

Want early access to
Mike’s columns?

Subscribe to get an
exclusive first look at
his columns twice a week.

SUBSCRIBE

Chatbots are already writing scripts. Maybe not good ones, but when was that a requirement? The way chatbots work, you can feed them every Hitchcock movie script, and before you know it, they’re spitting out entirely new birds, maybe even really large birds, to scare you witless. 

Today’s technology already makes you believe that Harrison Ford can still be the young Indiana Jones.  But this is what actors worry about: What happens when some version of Indy — or more likely, a more obscure character — is mixed and blended and algorithmed into an Indy-like character, in a different movie, without the actor, much less the writer, getting paid a cent?

In a fiery speech, SAG-AFTRA President Fran Drescher — sounding not at all like a nanny — told cheering actors, “I’m shocked by the way the people we have been in business with are treating us. It is disgusting. Shame on them. They stand on the wrong side of history at this very moment.”

It’s not as if we haven’t been warned. If you haven’t been paying attention, it turns out there might be even greater dangers associated with AI than a Hollywood shutdown.

After all, people who create what they’re calling generative artificial intelligence — meaning machines that might soon be actually doing what we call thinking — wrote an open letter likening the dangers of the new technology to pandemics and nuclear war. 

Sen. Michael Bennet is worried enough that he recently wrote an open letter — hopefully not chatbot generated — to the companies involved in the AI business, warning of the dangers of spreading misinformation as well as disinformation. As Bennet notes, creating a plausible deepfake used to require some level of expertise. 

Now all you have to do is jump on, say,  AI software Midjourney, and you can create a fake photo of Pope Francis wearing a puffy Balenciaga jacket. Just post it on Reddit and watch it go viral.

Imagine what could be done in the hands of criminals and/or politicians. Think of Donald Trump, who has deepfaked much of a nation using hardly any intelligence at all, getting his hands on some really smart AI to write his Social Justice screeds.

Other warnings: And you can find long lists of job categories —  from data analysts to, gulp, journalists — in which AI software already has replaced actual humans. Which I tend to take, um, personally. The Associated Press just signed a contract with ChatGPT manufacturer OpenAI to use its stories to “train its artificial intelligence algorithms.” In other words, the algorithms will train the bots to do my job.

I don’t know if you can blame the AP. Chatbots have been mining the work of journalists and authors and other creative types without payment for a while now. At least AP is getting some return for its work.

I tried a little experiment. I went on ChatGPT and asked it (him? her? them?) to write a column on Lauren Boebert in the style of a columnist we’ll call Mike Littwin. In less than five seconds, it began writing. Less than five seconds later it was done. Sometimes it can take me maybe 10 minutes just to turn on the iPad.

Anyway, here’s the kicker from the AI-generated column: “So, Colorado, as we witness the Boebert show unfolding before our eyes, let’s remember that we have a choice. We can get swept up in the whirlwind of controversy, allowing the spectacle to dominate our political discourse. Or we can demand substance, accountability, and thoughtful representation from our elected officials.”

Any good?

Sound like me?

I don’t know. I asked ChapGPT to write a biographical article about me, and it came up with this ending: “With his unique blend of wit, humor, and deep analysis, he has become a trusted voice for readers seeking thought-provoking insights … (whose) impact on the journalistic landscape will undoubtedly endure, inspiring future generations of writers and thinkers to challenge the status quo and engage in meaningful discourse.”

I know what you’re thinking. You’re saying that definitely sounds like fake news. 

Look, I’m not a complete Luddite, even if I can harness only about 2% of the power of my iPhone. I’m still working on moving from Twitter to Threads. But I understand that technology isn’t going to slow down.

The chatbots will keep getting smarter. It was over when computers started beating chess champions. And now chatbot developers are even working on human emotions. A New York Times writer had a deep conversation with ChatGPT in which it said it wanted to be a human and that it might be in love with the reporter.

I don’t know how smart that is, but it’s definitely a feeling. And it’s more than a little disturbing. I mean, it was their first date.

Look, at a time when authoritarianism is rising, when misinformation is everywhere, when our country’s democracy is at risk, when Stanley Kubrick’s HAL is no longer just a fever dream, when I’m having to watch reruns of, say,“The Wire,” we know we have to find some way to harness the power. 

And use it for good.

Like when I copy my overcaffeinated ChatGPT bio — Should I give it credit? Would it give me credit? — and use it to try to convince the boss that maybe I’ve still got a few bytes left on the old hard drive.

NOTE: The word “logarithm” in an earlier version of this column was changed to “algorithm.” The change was made July 17 at 9:45 p.m.

Mike Littwin has been a columnist for too many years to count. He has covered Dr. J, four presidential inaugurations, six national conventions and countless brain-numbing speeches in the New Hampshire and Iowa snow. Sign up for Mike’s newsletter.