The First lady says AI is the future of publishing. It’s already happening

U.S. first lady Melania Trump looks on during the unveiling of a Postal Service stamp honoring former first lady Barbara Bush in the East Room at the White House. (Kayla Bartkowski/Getty Images via CNN Newsource)

First lady Melania Trump released an audiobook version in her voice of her memoir on Thursday — but she won’t actually be the one narrating it.

“I am honored to bring you Melania – The AI Audiobook – narrated entirely using artificial intelligence in my own voice,” she wrote in a post on X. “Let the future of publishing begin.”

Trump is far from the first person to use AI this way. But her choice to put the technology and its use in media creation on a bigger stage hints at the bigger role AI may soon play in creating everything from the news articles people read to the videos and shows they watch — and raising questions about whether media jobs will survive the change.

“It’s too reductive to say, yes, that’s an inevitable cut in the number of jobs,” Alex Connock, senior fellow in management practice at the University of Oxford‘s Saïd Business School, told CNN. “But it’s also fanciful to say there’s going to be no change to how employment works.”

(Not) coming soon to a theater near you

Trump’s book will be narrated by an AI-generated copy of her voice that was “created under Mrs. Trump’s direction and supervision,” the product description on her website reads.

Experts say using AI for voiceover work is becoming common, especially as tech from companies like Google and ElevenLabs — the firm Trump used to create her AI audiobook — make it easy to turn text-based materials into audio that sounds like a podcast.

But Trump’s announcement brought that AI use to the fore.

“I don’t think that there’s going to be a rush to (an) immediate replacement of voiceover,” Clay Shirky, vice provost for AI and technology in education at New York University, told CNN. “A lot of these things happen gradually, but it certainly is a milestone.”

The Trump audiobook comes as tech giants are launching tools that make it increasingly easy for anyone to generate realistic video and audio with little effort.

Within the same week Trump announced her audiobook, Google debuted a more advanced version of its video generation model that can create audio — even dialogue between characters — to match the scene.

Late last year, OpenAI released a video creation tool called Sora, which was so popular the company had to temporarily pause signups because of high demand. The ChatGPT maker ran into a similar issue earlier this year when its image generation tool went viral for its ability to create pictures resembling the style of Japanese animation company Studio Ghibli.

But that doesn’t mean AI-generated feature films are coming anytime soon. The current version of the technology is ideal for creating short form videos you might find on social media, according to Shirky.

What’s more likely is that TV networks and production companies will look for new ways to incorporate AI into existing programs. Connock, who consults with TV production companies, says he’s had multiple meetings this week alone with those in the TV industry looking to learn more about AI, which he says is a major change from a year ago.

Connock says producers are curious about creating AI replicas of TV personalities that viewers can interact with while watching their show. He attributes the increased interest to a desire to keep up with social media creators.

“The kind of old school, traditional professional TV economy has realized that in order to even compete at all with the creators, they have to at least match them shot for shot on their ability to deploy those tools,” he said.

AI could enable a shift from media that’s meant to be watched or read to digital content that viewers can interact with, according to Oren Etzioni, former CEO of the Allen Institute for Artificial Intelligence and professor emeritus at the University of Washington.

“What if you could actually talk to Melania Trump about the chapter?” he told CNN. “That’s coming soon, maybe not with her, but you know, coming soon to a book near you.”

AI and the future of jobs

The launch of Trump’s audiobook also comes as AI-generated content has raised questions about whether AI will take humans’ jobs as it gets better at tasks like creating podcasts, authoring books and writing code.

The World Economic Forum’s Future of Jobs report, released earlier this year, found that 41 per cent of employers plan to downsize as generative AI plays a bigger role in work-related tasks. Aneesh Raman, the chief economic opportunity officer at LinkedIn, recently fretted about AI replacing some entry-level jobs in a New York Times op-ed.

Those fears have been especially prevalent in the media industry; film and TV writers in the Writers Guild of America went on strike in 2023 in part to prevent aspects of their jobs from being replaced with AI. An agreement was reached after 146 days establishing that AI can’t be used to “write or rewrite literary material.”

The WGA did not respond to CNN’s request for comment.

But answering the question of whether AI will replace media jobs is complicated; experts see some areas, like voiceover work, that could be impacted quickly. Yet other roles that involve nuanced handling of sensitive data will be more challenging to fill with AI.

“If I’m an investigative journalist, and I spend a lot of time getting to know people and understanding complex situations, that’s not a job that’s easy to replace,” said Shirky.

It’s also possible that the answer will fall somewhere in between; companies may shift their hiring practices to include professionals with AI expertise.

But that might not mean job cuts.

“Traditionally, a development department would be kind of three people with arts degrees,” said Connock. “And now it might be one person with an arts degree, one person who’s a kind of professional coder, and one person who’s kind of an academic researcher.”

Article written by Lisa Eadicicco, CNN