AI avatars of deceased people — often called “deadbots” — are emerging in emotionally charged settings, from advocacy interviews to courtroom videos, prompting researchers and industry figures to discuss their persuasive power and potential commercialization. According to NPR, families and companies are experimenting with voice- and video-based recreations that can speak in the likeness of the dead.
Persuasion and growing interest
Recent examples highlight how these avatars can influence audiences. The family of Joaquin Oliver, who was killed in the 2018 Parkland school shooting, created a beanie-wearing AI avatar that spoke with journalist Jim Acosta in July. “This is just another advocacy tool to create that urgency of making things change,” Manuel Oliver told NPR. In May, an AI avatar of Chris Pelkey, a road rage victim in Arizona, delivered a video impact statement at the shooter’s sentencing; Judge Todd Lang called it “genuine” before issuing the maximum sentence.
The digital afterlife industry, which includes deadbots, is expected to quadruple to nearly $80 billion over the next decade, NPR reports, citing market research. Cartoonist Amy Kurzweil, whose 2023 book “Artificial: A Love Story” recounts building a text-based bot of her late grandfather with Ray Kurzweil, described the medium’s pull: “I could feel like I had some communion with his presence.” She added that richer voice and video avatars can be “more affecting.”
“Primed for monetization”
Lindenwood University researcher James Hutson said, “Of course it will be monetized,” referencing studies on ethics and commercialization, including work from Cambridge University exploring deadbots used to advertise. Hutson noted Americans have long seen deceased figures in ads, pointing to 1990s commercials featuring a digitally manipulated Fred Astaire, and observed that audiences have accepted ads even in some paid streaming tiers.
Legal and ethical questions
Attorney Jeffrey Rosenthal of Blank Rome LLP said U.S. law lacks comprehensive federal AI rules, and existing state right-of-publicity statutes may be ill-equipped for scenarios like a deadbot relative delivering ads. “Who is liable?” he asked, pointing to possible responsibility across software makers, advertisers, and IP owners.
Industry voices expressed discomfort with using deadbots to sell. “Ethically, I think using dead people is not sound at all,” said Camille Chiang of AI marketing firm NEX. Alex Quinn, CEO of Authentic Interactions Incorporated, parent of StoryFile, said it would “turn me off as a consumer,” but he is “absolutely interested” in ad-friendly formats such as interstitial breaks during deadbot conversations or having avatars probe for user preferences. Quinn said companies are already testing internal use cases, though implementations remain limited, NPR reports.
