Summary: What happens when artificial intelligence makes the dead speak again? A new, wide-ranging study examines this unsettling frontier of digital resurrection. By analyzing more than 50 real-world cases, researchers show how generative AI is transforming the voices, faces, and life stories of the deceased into reusable data and what the authors call “spectral labor.”
From hologram concerts to grief-focused chatbots, these technologies are redrawing the line between life and death. They are producing a form of “digital afterlife” in which the dead are repurposed for profit, politics, and personal comfort—often without prior consent.
Key Facts
- Spectral Labor: A term introduced by the study to describe how the dead are effectively put to work by supplying data—voice, image, text—that serves the living’s emotional, political, or commercial aims.
- Three Modes of Resurrection:
- Spectacularization: AI-generated performances that revive public figures and icons for entertainment.
- Sociopoliticization: Reanimating victims of injustice to testify or appear in political or commemorative contexts.
- Mundanization: Everyday people using chatbots or synthetic media to maintain ongoing interactions with deceased loved ones.
- Consent Vacuum: Most AI-driven resurrections occur without clear ownership rules, explicit consent from the deceased, or robust accountability measures.
- Postmortal Society: Society is shifting toward seeking continuity after death through data and algorithms, creating digitally sustained “afterlives.”
- Ideological Weaponization: AI can extend the influence of political figures and ideologues after death, allowing their messages to circulate indefinitely and without oversight.
Source: Hebrew University of Jerusalem
Overview: Generative AI is already being used to “bring back” the dead as entertainment figures, political witnesses, and companions for grieving families. The study argues this is not only emotionally powerful but also ethically fraught, because AI treats a person’s voice, likeness, and memories as raw material to be reused and monetized.
Because these revivals often happen with little oversight or consent, they create a novel form of exploitation. The authors use the phrase “spectral labor” to capture how the dead are turned into involuntary sources of data and profit. Families and societies must now navigate blurred lines between memory and manipulation, comfort and coercion, tribute and misuse.
What does it mean when artificial intelligence makes the dead speak again? The study takes examples ranging from hologram concerts of long-deceased stars to chatbots trained on the texts and recordings of loved ones. It highlights how generative AI is rapidly reshaping the boundary between living relationships and digital continuations.
Conducted by Tom Divon (Hebrew University) and Prof. Christian Pentzold (Leipzig University), the research—titled Artificially Alive: An Exploration of AI Resurrections and Spectral Labor Modes in a Postmortal Society—reviews over 50 cases across the United States, Europe, the Middle East, and East Asia. Rather than focusing on a single viral example, the authors map a cross-cultural pattern that illustrates how AI resurrections are becoming normalized and institutionalized.
Their analysis identifies three main modes by which the dead are digitally reintegrated into public and private life: celebrity spectacles, political or commemorative uses, and intimate, everyday interactions. Across all three, the same dynamic appears: the extraction, repackaging, and repeated use of a deceased person’s data for various ends—often without explicit authorization.
Three ways AI reintroduces the dead
- Spectacularization – Public, entertainment-driven recreations of famous figures. AI can produce “new” performances or immersive shows featuring iconic artists.
- Sociopoliticization – Using AI to re-present victims or historical figures in political, educational, or commemorative settings, sometimes to testify or make political claims posthumously.
- Mundanization – Personal and intimate uses, where chatbots and synthetic media recreate the voices and personalities of lost family members for ongoing interaction.
Spectral labor and the ethics of posthumous use
Divon and Pentzold show how AI systems are trained on the digital remains of people—photos, recordings, messages, and profiles—and then used for profit, persuasion, or consolation. Without consent, these data are extracted and monetized, creating risks of exploitation and ideological misuse. The authors warn of scenarios in which a political figure’s voice continues to spread ideology after death or where a victim’s likeness is repeatedly used in ways that resurface trauma.
The central ethical questions include: who owns a voice after death, who controls a digital likeness, and who decides how, when, or why a person should be digitally revived? These questions demand legal and social responses before such practices become routine.
Living in a postmortal society
The study frames these developments within the idea of a postmortal society—one where death is not denied but increasingly managed through technological means. In this environment, data and algorithms promise a kind of continuity beyond biological life. Yet the authors emphasize that AI does not end death; it suspends people in a troubling in-between state, neither fully alive nor fully gone.
As generative AI proliferates, the researchers call for urgent ethical and legal discussion to prevent unregulated normalization of digital resurrection. “Thinking seriously about what AI does to our relationship with the dead,” they write, “is essential to understanding what it is doing to the living.”
Key Questions Answered:
A: Technically, no. The study argues, however, that these AI recreations create a lived experience for others—a state of “re-presencing” that can feel emotionally real and alter how people relate to the deceased.
A: The technology to create such chatbots exists and is already in use. The study warns that this is happening in a legal and ethical “Wild West” with few clear rules about ownership or consent.
A: Exploitation. The dead cannot consent or refuse, making their likenesses and data vulnerable to commercial, political, or emotional exploitation.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The journal paper was reviewed in full by the editorial team.
- Additional explanatory context was added by staff to clarify findings and implications.
About this AI and neuroethics research news
Author: Press Office
Source: Hebrew University of Jerusalem
Contact: Press Office – Hebrew University of Jerusalem
Image: Image credited to Neuroscience News
Original Research: Open access. “Artificially alive: An exploration of AI resurrections and spectral labor modes in a postmortal society” by Tom Divon and Christian Pentzold. New Media & Society. DOI: 10.1177/14614448251397518
Abstract
Artificially alive: An exploration of AI resurrections and spectral labor modes in a postmortal society
Generative artificial intelligence is changing how we understand life and death. One experimental application is the recreation of deceased human beings. This article examines that nascent field, analyzing 50 cases across multiple regions and identifying three principal modes of AI resurrection: (1) Spectacularization—public, entertainment-focused recreations of iconic figures; (2) Sociopoliticization—the use of recreated personas in political or commemorative contexts; and (3) Mundanization—the everyday revival of loved ones through chatbots and synthetic media.
To describe the industry dynamics that monetize digital remains, the authors introduce the concept of spectral labor: the extraction, circulation, and monetization of a deceased person’s data, likeness, and affect, often without consent. The paper argues that these uses of generative AI raise urgent legal and ethical concerns about posthumous appropriation, ownership, and control.