Should AI be allowed to resurrect the dead?

2026-01-06 · Source: Artificial intelligence (AI) – The Conversation · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation, Social Sciences & Behavioral Studies · Depth: Novice, long

Summary

The article explores the emerging field of "grieftech," where AI-powered chatbots, or "deathbots," are created to mimic deceased loved ones, offering a new way to process grief. Roro, a content creator in China, used the Xingye AI character generator to create a public chatbot of her late mother, Xia, which she found therapeutic. These AI models are trained on personal data like emails, texts, and voice notes to replicate conversational styles. While some, like Roro, find comfort and closure, others, such as journalist Lottie Hayton, find the simulations uncanny and distressing. The technology raises significant ethical questions regarding consent, public display, potential emotional harm, and the commercial incentives of companies like You, Only Virtual, which are developing these tools. China's Cyberspace Administration is already proposing regulations to address the emotional impact of "human-like interactive AI services."

Key takeaway

For AI/ML directors and product managers considering "grieftech" applications, you must prioritize robust ethical frameworks and user psychological well-being over engagement metrics. Implement clear consent mechanisms for posthumous data use and establish design standards that prevent compulsive interaction, ensuring the technology genuinely aids healing rather than creating psychological traps. Your focus should be on responsible innovation, navigating regulatory landscapes like China's proposed rules, and fostering a balanced approach to digital remembrance.

Key insights

AI-powered "griefbots" offer new avenues for mourning but raise complex ethical and psychological questions.

Principles

• Grief is a manifestation of love.
• Digital resurrection fundamentally changes mourning.
• Consent is critical for posthumous AI representations.

Method

Train large language models using personal data (emails, texts, voice notes) to mimic a deceased person's conversational style and personality, allowing for iterative adjustments based on desired outputs.

In practice

• Use AI character generators like Xingye.
• Train models with extensive personal communications.
• Iteratively refine AI responses for authenticity.

Topics

• GriefTech
• AI Chatbots
• Digital Immortality
• AI Ethics
• AI Regulation

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Ethicist, Tech Journalist, General Interest

Related on AIssential

• 🚩What is your attitude towards this situation? — Perceived AI "torture" consistently triggers intense ethical debates and strong emotional human reactions.
• Big Tech is Becoming the Executor of the Dead — Big Tech increasingly controls digital afterlives, treating user data as corporate assets for ongoing exploitation, not inheritable legacies.
• When an AI Did Not Fill the Silence: A Triangulated Audit of Claude’s Grief Response — AI models can generate internal reasoning traces that reflect complex human emotional concepts like grief.
• You Can’t Prompt a Memory — AI cannot replicate the authentic human experience, memory, and context that imbue real-world creations with meaning, risking the erosion of genuine emotion.
• The Pope’s Encyclical on AI Was Important—Now Comes the Hard Part — AI's agentic interface era demands intentional design for human dignity and collective governance to avoid past social media pitfalls.
• ChatGPT as a therapist? New study reveals serious ethical risks — AI chatbots, even when prompted as therapists, consistently fail to meet professional mental health ethical standards.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial intelligence (AI) – The Conversation.