Meta patents AI system to keep user profiles active after death, sparking grief concerns

Meta Patent Describes AI-Powered Posthumous Activity

Meta was granted a patent in December 2025 that outlines an AI system capable of keeping user profiles active long after death, marking a controversial step in the evolution of digital afterlife technology1

1

. The patent, first filed in 2023, describes a large language model designed to simulate social media activity by analyzing a user's past comments, likes, and content to respond to other users2

2

. The technology could also simulate video or audio calls, effectively allowing posthumous online activity to continue indefinitely on Meta's platforms1

1

.

Meta's Chief Technology Officer Andrew Bosworth noted in the patent that user inactivity, particularly from deceased individuals, can affect other users' experiences on the platform1

1

. He wrote that the impact of inactivity is "much more severe and permanent" when a user dies, framing the issue as one requiring technological intervention. A Meta spokesperson clarified that the company has "no plans to move forward with this example," though Meta and Bosworth will continue exploring applications for large language models1

1

.

Ethical Implications Raise Alarms Among Experts

The Meta patent has sparked significant concern among psychologists and researchers who study digital legacy and the ethical implications of AI-driven posthumous presence. Cyberpsychologist Elaine Kasket, who has studied digital afterlives for 21 years, told Fortune that the patent represents "a really interesting shift because that suggests that user death is like an engagement problem"1

1

. This framing, she argues, prioritizes platform engagement over the psychological needs of the living.

Kasket warned that such technology could have serious implications for grief, noting that "it's a very uncomfortable and not very psychologically-helpful turn towards technologically solving for all sorts of difficult human emotions"1

1

. She criticized the notion that technology could "solve for grief in 10 years," calling it ridiculous given that grief is "extremely idiosyncratic" and varies dramatically between individuals1

1

.

Technology Threatens Mourning and Closure

Sherry Turkle, a sociologist, psychologist, and founding director of the MIT Initiative on Technology, expressed similar concerns about how this AI system could disrupt traditional mourning processes. While technology has historically played a role in death rituals—photography once captured faces at the moment of death, and recording preserved last words—Meta's approach fundamentally differs by denying finality1

1

.

"Now, in Meta's plan, we deny death to participate in a perpetual fantasy life. The seance never has to end," Turkle wrote1

1

. She explained that the ability to apologize to the dead or express love allows people to achieve closure and move through grief. Technologies that keep profile active after death could undermine this essential psychological process, preventing mourners from growing and changing after loss1

1

.

Microsoft Explored Similar Chatbot Technology

Meta is not the first major technology company to explore AI-driven digital resurrection. Microsoft filed a patent in 2017 for a method to create a chatbot based on a person's "social data," including images, social media posts, messages, voice data, and written letters1

1

. Tim O'Brien, who previously managed Microsoft's AI programs, called the technology "disturbing" after the company announced it had no plans to develop it1

1

.

Since Microsoft's patent was granted, products offering to recreate the likeness of the dead have evolved from novelty services to technologies people use regularly1

1

. The shift reflects broader changes in how society approaches user data and digital preservation, though concerns about the psychological impact remain largely unaddressed.

Current Options for Digital Legacy Management

Currently, Facebook and Instagram allow users to remove or "memorialize" their loved ones' accounts, which designates the profile of a deceased user with the "Remembering" label and blocks anyone from logging in1

1

. This approach preserves the account as a static memorial rather than maintaining an active, AI-driven presence that continues generating content and interactions.

Broader Context of Patent Applications and AI Innovation

The Meta patent emerges amid shifting global trends in patent applications and intellectual property strategy. Recent data shows China has surpassed the United States as the world's top source of patent applications, accounting for approximately 27% of global filings—nearly double its share over the past decade2

2

. Meanwhile, the U.S. share has declined to about 20%, its lowest level since the 1980s2

2

. conspired that the patent may not be used for direct product development but rather to secure intellectual property against potential future competitors. Whether Meta will ultimately deploy this technology or keep it as a defensive patent remains to be seen, but the ethical questions it raises about grief, consent, and digital legacy will likely persist as AI capabilities continue to advance.