Scott Adams once sounded open to the idea of a digital afterlife. Now that he’s passed, social media posts attributed to his family say an AI version of the “Dilbert” creator circulating online is unauthorized — and deeply distressing.
In a 2021 podcast clip, the cartoonist said he granted “explicit permission” for anyone to make a posthumous AI based on him, arguing that his public thoughts and words are “so pervasive on the internet” that he’d be “a good candidate to turn into AI.” He added that he was OK with an AI version of him saying new things after he died, as long as they seemed compatible with what he might say while alive.
Shortly after the 68-year-old’s January death from complications of metastatic prostate cancer, an AI-generated “Scott Adams” account began posting videos of a digital version of the cartoonist speaking directly to viewers about current events and philosophy, mirroring the cadence and topics the actual human Adams discussed for years.
His family says it’s a violation, not a tribute.
A February 5 post on Adams’ official account attributed to his brother, Dave Adams, insisted the cartoonist “never intended, never would have approved an AI version of him that wasn’t authorized by himself or his estate.”
“The real Scott Adams gave explicit permission on the record multiple times for people to create and operate an AI version of him,” the AI Adams said in a post on February 5. “So this iteration exists as a direct fulfillment of that stated wish.”
The official Adams account reiterated the family’s objection on February 17, saying the estate was “kindly but firmly” asking anyone using AI to recreate his voice or likeness to stop, calling the digital replicas a “fabricated version” of Adams that is “deeply distressing.”
“This is not a tribute. It is not an honor. It is an unauthorized use of identity,” the post read.
The Adams estate and the AI Adams account did not respond to requests for comment from Business Insider
The dispute underscores the growing legal and ethical fault lines around “AI afterlives” — and how quickly technology can outpace the rules meant to govern it.
‘It’s a deepfake’
Karen North, a University of Southern California professor specializing in digital social media and psychology, said calling the AI-generated Adams an avatar, as some have online, softens what it is.
“It’s a deepfake,” North told Business Insider.
The troubling part, she said, is how a realistic imitation can surface while a family is grieving and potentially say things the real person never would have said. North added that since many Americans are “giving up so much information” through apps that capture faces and voices and viral quizzes that collect personal details, it is increasingly easy to recreate someone without permission.
“I find it very disturbing,” she said.
Betsy Rosenblatt, an intellectual property lawyer and professor at Case Western Reserve University, said her initial reaction was that the AI Adams is “unethical in the extreme.”
“When people die, they die,” she said.
Legally, she said, the central issue is the right of publicity — protections over a person’s name, image, and likeness. Still, those laws are more focused on privacy and economics than on grief.
The right of publicity is “chiefly concerned with economic remedies,” Rosenblatt said.
The strongest claims typically involve money: an AI version could harm existing deals tied to Adams’ identity or block the family from striking their own.
Rosenblatt described two potential economic harms: “One is that it could be harming some financial arrangement that they already have. Another is that it might stand in the way of their making some competitive financial arrangement,” she said.
The account appears to be anonymous; however, that wouldn’t necessarily prevent a lawsuit.
“You can sue somebody who is anonymous,” Rosenblatt said, and courts can allow subpoenas to uncover identifying information, though it’s “not necessarily easy.”
The legal analysis also hinges on whether the account is commercial. Courts often ask whether the speech proposes a commercial transaction.
If the digital replica isn’t selling anything, Rosenblatt said, it becomes “more likely to be considered a First Amendment protected expression” for the anonymous creator — not a “slam dunk,” but a stronger argument.
The AI Adams identifies itself as artificial intelligence at the start of its clips and does not appear to solicit money.
In a February 1 post, it said: “The original Scott’s gone, passed on, but the thinking survives.”
Consent isn’t the same as a contract
The estate’s objections sit uneasily alongside Adams’ 2021 comments offering “explicit permission” for AI versions of him.
North said offhand remarks about technology shouldn’t automatically be treated as binding authorization. Adams was “an incredibly bright, incredibly creative person” who often pushed boundaries, she said, and comments made in conversation “may not be legally binding in ways contracts and intellectual property rights are legally binding.”
“Let this be a warning to all of us: be careful what you say, because he’s now put his loved ones in a difficult position as they protect his legacy,” North said.
Rosenblatt said Adams’ wishes “would certainly matter in an ethical sense,” but may not matter legally “unless he gave somebody the legal rights to do that.”
There is no comprehensive federal law governing posthumous AI likeness, but some states — like New York and California — have recently enacted laws requiring consent from heirs or estate executors before creating digital replicas.
Beyond legal questions lies a deeper ethical one: who controls a person’s persona after they’re gone?
North said people “should own the rights to our own personas,” and when they die, those rights “should go to our loved ones,” not become a free-for-all. AI replicas, she warned, can drift off-brand or reshape public memory.
“Shakespeare should always sound like Shakespeare,” she said. “Dr. Seuss should always sound like Dr. Seuss.”
For now, the AI “Scott Adams” fight is one family’s public line-drawing exercise. It may also be a preview of a broader reckoning in a world where convincing digital imitations are easy to make — and where the law is still struggling to answer who gets to decide whether the dead keep talking online.
Read the full article here
