Postmortem avatars in grief therapy: Prospects, ethics, and governance
This paper explores the ethical viability and therapeutic potential of using postmortem AI avatars in grief counseling, proposing specific clinical applications while arguing that ethical concerns can be mitigated within a therapeutic context and emphasizing the urgent need for empirical research and governance frameworks.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you have lost someone you love. The silence they leave behind is deafening. Now, imagine a new technology that can mimic their voice, their jokes, and their way of speaking, allowing you to have a conversation with them again. This is what Postmortem Avatars (PMAs) are: AI chatbots trained on a deceased person's old texts, emails, and voice notes to sound like them.
While these "digital ghosts" are becoming popular, they are mostly sold as consumer products (like a cool app you buy on your phone). This paper asks a big question: Could these avatars be used safely and effectively by doctors and therapists to help people heal from grief?
Here is a simple breakdown of the paper's main ideas, using some everyday analogies.
1. The Core Idea: Grief is a Dance, Not a Stop Sign
The authors start by explaining that grief isn't just about being sad; it's a process. Think of it like a dance with two steps:
- Step A (Looking Back): Feeling the pain, missing the person, and talking about the loss.
- Step B (Moving Forward): Learning to live without them, finding new routines, and rebuilding your life.
Healthy grieving is dancing back and forth between these two steps. If you get stuck only doing Step A, you might get "stuck" in grief (Prolonged Grief Disorder).
The paper suggests that PMAs could actually help with this dance. Instead of blocking the "looking back" step, a therapist could use a PMA to help you say the things you never got to say, or to help you practice letting go in a safe environment.
2. The "Magic Trick" of Make-Believe
A major worry is: Won't people get tricked into thinking the dead person is actually alive?
The authors say: No, and that's okay. They use a concept called Fictionalism.
- The Analogy: Think of a PMA like a very good actor in a play. When you watch a play, you know the actor isn't really the character. But you still cry, you still feel the emotion, and you get swept up in the story. You are playing along with a "game of make-believe."
- The Point: Most people using these avatars know it's a computer program. They aren't deluded; they are choosing to engage in a "pretend" conversation to help process their feelings. It's like talking to an empty chair in therapy, but the chair has a voice.
3. Two Ways Therapists Could Use This
The paper suggests two specific ways a therapist (a "guide") could use these avatars, rather than just letting people use them alone:
- The "High-Tech Empty Chair": In traditional therapy, a patient sits across from an empty chair and imagines their loved one is sitting there. It's powerful, but it relies entirely on the patient's imagination.
- The PMA Twist: The therapist could use an AI avatar to fill that chair. It can say things the patient didn't expect, helping them see their relationship from a new angle. It's like upgrading from a black-and-white sketch to a full-color movie, but still keeping the safety of the therapy room.
- The "Digital Scrapbook": Usually, grief therapy involves drawing a picture of the deceased or making a memory box.
- The PMA Twist: The process of building the avatar could be the therapy. The patient has to decide: "What jokes should the bot know?" "How should it sound?" "What stories should it tell?" This forces the patient to actively remember and curate their loved one's memory, which is a healing act in itself.
4. The Five Big Worries (and Why They Might Be Manageable)
Critics have five main fears. The authors argue that in a clinical setting (with a therapist present), these fears are less scary than they seem:
- "It stops you from grieving."
- Counter: We use medicine (like antidepressants) to help us feel better even though it changes our "natural" emotions. If a PMA helps someone move through grief, is it really "unnatural"?
- "It deceives people."
- Counter: In therapy, feeling the "reality" of the conversation is the point. Just as a placebo pill can heal because you believe it works, a PMA can heal because it feels real enough to trigger emotional release. Plus, therapists are trained to spot if someone is losing touch with reality.
- "It's addictive."
- Counter: People can get addicted to many things (even medicine). The solution isn't to ban the tool, but to regulate it. A therapist acts like a doctor prescribing a drug: they monitor the dose and stop it if it becomes harmful.
- "It violates the dead person's rights."
- Counter: Actors play dead people all the time and take creative liberties. If the goal is healing the living, the avatar doesn't need to be a perfect, 100% accurate replica. It just needs to be a helpful representation.
- "Why not just use Deepfakes?"
- Counter: Deepfakes are usually videos made by a therapist. PMAs are different because the patient helps build the avatar. It's the difference between a therapist painting a portrait of your dad for you, versus you painting your dad yourself. The act of creating it is part of the healing.
5. The Missing Piece: We Need More Research
The authors are very careful. They are not saying: "Go buy a PMA today and use it in therapy."
They are saying: "Don't ban it yet. We don't have enough data."
Right now, we are flying blind. We don't know if these tools help or hurt in the long run. The paper calls for a "test drive" phase where researchers, doctors, and AI developers work together to:
- Test if it actually helps people heal.
- Create rules (governance) so that only trained professionals can use them for therapy.
- Make sure the "digital ghosts" don't become a trap.
The Bottom Line
Think of Postmortem Avatars as a new, powerful tool in a toolbox.
Right now, the tool is a bit rusty and dangerous if used by a novice. But the authors argue that if a skilled mechanic (a therapist) uses it carefully, it might be able to fix a broken engine (grief) that other tools can't reach.
The goal isn't to bring the dead back to life, but to help the living find a way to carry their love forward without being crushed by the weight of the loss.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.