Death has always signified the end. But for tech companies, it might be the start of another business.
At the end of last year, Meta patented an AI that could post on behalf of a person even after they’re gone. In November 2023, Meta's CTO Andrew Bosworth filed a patent called “Simulation of a user of a social networking system using a language model” (“Simulation of a user of a social networking system using a language model”).
After two years of evaluation, the process was finished: on December 30, 2025, Meta officially received this patent.
“A language model can be used to simulate a user when the user is absent from the social networking system, such as when the user takes a long break or if the user dies,” states the patent.
The idea of digital immortality isn’t new for Meta. The company previously offered a “legacy contact” option that allowed surviving relatives to manage deceased users' accounts. During the metaverse era in 2023, Mark Zuckerberg hinted in an interview with podcaster Lex Fridman that his goal was to create virtual avatars for the deceased, enabling them to interact with living users posthumously.
“If someone has lost a loved one and is grieving, it can be helpful to communicate with them or relive certain memories,” Zuckerberg said at the time.
To make the simulation believable, the system won't just rely on publicly available posts. The algorithm is set to analyze the complete digital footprint: from private messages and like history to voice messages. The model would send not just generic messages but mimic a person’s specific tone, sense of humor, and personal preferences.
Even though a Meta spokesperson told Business Insider that the company has no intention of pursuing this further, the mere existence of the patent and its approval suggests that the technological groundwork is in place. Even if Meta never launches this specific tool, academics are already critically engaging with the topic.
The Commercialization of Death
Integrating AI into posthumous digital life could transform the very nature of human grieving. Grief, one of the most intimate psychological processes, could become a commodity. Experts from Cambridge University call this “the digital posthumous industry” and warned about it as early as 2024.
The authors worked with a hypothetical scenario where a service called MaNana allows users to communicate with deceased's digital versions (“deadbots”), created from analyses of old voice and text messages.
As a hypothetical example, they mentioned 35-year-old Bianca, who lost her grandmother. Bianca can't afford the 50 euro monthly subscription, so she opts for a free version, but the price for this “free eternity” is high, as the simulation of her late grandmother Laura is interrupted by ads. When Bianca calls her digital grandmother for a family carbonara recipe, the AI interrupts, suggesting she order it from a specific delivery service instead.
When Bianca decides to end the simulation out of emotional turmoil, she discovers the app lacks any mechanism for dignified “shutdown” or deletion of the bot itself. Her late grandmother thus remains in digital limbo, a puppet in corporate hands with no escape.
“People can form strong emotional attachments to such simulations, making them particularly vulnerable to manipulation,” said study co-author Dr. Tomasz Hollanek from LCFI, Cambridge.
If interaction with the deceased is governed by a functional algorithm, nostalgia and emotional bonds to loved ones become a business transaction. Death becomes a simple profit source. And while wealthier users might pay for a “dignified” ad-free version of digital immortality, for others, memories become just another ad space.
In this sense, digital posthumous life could represent an extreme form of so-called technofeudalism, a concept defined by Greek economist and former Finance Minister Yanis Varoufakis.
User data, collected during their lifetime, could remain a profit source even after their death. Sadly, this “digital estate” might not release a person - because they're dead. At least biologically.
Postmortem Privacy
The problem isn't just annoying ads or tech giants chasing profit. What happens to human dignity? Experts in the Cambridge study refer to “postmortem privacy,” which in 2017 was defined by academic Edina Harbinja as “the right of a person to protect and decide what happens to their reputation, dignity, integrity, secrets, or legacy after death.” If corporations use digital remnants to covertly influence the purchasing behavior of survivors, there's a fundamental breach of user autonomy and data donor dignity. The deceased become mere ad carriers, powerless to fight it.
Moreover, authenticity becomes an issue. A digital version of a deceased person is a synthetic model generating content the person never actually expressed. Identities become algorithmic products, not memories. And do we even know if the dead would want to keep “living” in this way posthumously? If so, wouldn’t they prefer their memory be preserved naturally? While this option may help the grieving in the short term, would they really want their loved one to become a product?
What Happens to Grieving?
Grieving is a crucial psychological process everyone must go through at least once. It's a moment where we learn to accept loss, process grief, and gradually adjust to a world without our loved one. But digital simulations of the deceased could fundamentally disrupt this process. Instead of natural closure, survivors end up in endless algorithmic interaction, complicating emotional loss processing and potentially delegitimizing the deceased's life.
Survivors are often strongly emotionally impacted by the death of a loved one and may be unable to think entirely rationally. Interacting with digital versions simulating their voice and personality can amplify this vulnerability, leading to reality confusion or psychological dependence - a phenomenon some psychology and psychiatry experts term in popular literature as “AI psychosis”.
Grieftech Instead of Grieving?
This phenomenon isn’t entirely new and already has a name: grieftech. Even if Meta's patent never goes live, platforms like HereAfter AI or Project December already let people upload memories and voices in life. The result? Interactive avatars that can “speak” at funerals or answer questions from grandchildren.
We can't escape death, but we can avoid supporting those who want to profit from it. Maybe it's best to stick to good old grieving.
