Los Angeles County’s motion picture and sound recording sector lost about 6,700 jobs in the twelve months through May 2026, according to data from California’s Employment Development Department. Those losses accounted for more than 90% of all employment declines across the county’s entire information industry over the same period.
That is not a monthly wobble. It is a sustained, year-long contraction concentrated almost entirely in one sector of one region.
The industry has offered several explanations, and most of them are partially true. Production volume is down sharply — by some estimates around 30%. Studios have been cutting overhead: Disney announced further layoffs at Pixar, National Geographic, and ESPN this month, following earlier reductions at Marvel. Work that remains is increasingly shifted to freelancers rather than staff.
Underneath all of it sits a technology question the industry argues about publicly and answers privately.
What the job postings show
A Los Angeles Times survey of several hundred studio job listings posted in late June found that more than one in ten was likely connected to artificial intelligence. The postings fall into two categories, and both are revealing.
The first is construction. Disney has been recruiting doctoral-level researchers to work on computer graphics and AI for Pixar and Disney films, and to move research into practical studio use. Industrial Light & Magic advertised for supervisors to explore machine learning in new production workflows. Skywalker Sound appears to be building proprietary models for soundtrack work, voice separation, and voice transfer.
The second is defense. Three of Disney’s listings were content-security roles focused on evaluating AI tools, guarding against piracy, and handling watermarking and rights protection.
The same companies suing over unauthorized use of their libraries are hiring aggressively to use the technology themselves. One executive quoted in the survey described studios that hire for AI constantly but never discuss it publicly.
The cost math driving it
The economics are not subtle. Short-form mobile series, which lean heavily on AI-assisted production, have grown into a business worth roughly $1.3 billion. Some production companies in that segment can produce a thirty-minute show for a few thousand dollars, against costs that would historically have run into the hundreds of thousands with human casts and crews.
Lionsgate has moved fastest among the traditional studios, naming Kathleen Grace as its first chief artificial intelligence officer and deploying AI tools across a large majority of its workforce. Netflix acquired an AI filmmaking startup founded by Ben Affleck to accelerate production and post-production work.
Whether the savings are real at scale is genuinely unsettled. Frontier AI tools carry meaningful compute costs, and pushing those costs onto freelancers rather than absorbing them in-house may simply relocate the expense rather than eliminate it.
Where the rules get written
The labor framework is being built in real time. After a lengthy strike, SAG-AFTRA ratified an interactive media agreement in 2025 requiring consent and disclosure for AI digital replicas, and allowing performers to suspend that consent during a strike — which effectively sends a performer’s digital double to the picket line alongside them.
The next round of negotiations between SAG-AFTRA and the Alliance of Motion Picture and Television Producers will set the floor on AI usage across film, television, and streaming. The terms that emerge will not stay in Hollywood. Advertising, gaming, and live events all take their cues from these contracts.
For the 6,700 people already counted in California’s data, those negotiations arrive after the fact. The employment numbers are not a forecast of what AI might do to a creative workforce. They are a record of what has already happened to one.
Source reporting: California Employment Development Department, Los Angeles Times, Forbes, industry filings.