Wanted Dead or Alive: Anticipatory Regulation of Digital Immortality Platforms

Abstract

The loss of a loved one is a painful and unique experience. To address that loss, technology companies are exploring the potential of digital immortality platforms, preserving individuals through artificial intelligence (AI) models trained on digital data generated in life. These datasets can be collected from scraping social media profiles and text conversations postmortem or intentionally recording videos and conversations with a person prior to their death for training AI models. In the face of any novel technology, the potential risks and benefits should guide the responsible development and deployment of every product. However, the liminal regulatory space that digital immortality technologies occupy makes this type of evaluation a difficult undertaking. One such approach would be to examine analogous sectors to the novel technology as a first step in developing regulations in that space. In this paper, we identify several ethical risks presented by digital immortality technologies and propose four analogies to contextualize possible regulation: psychology/grief counseling, death care/remains management, social media platforms, and AI companion services. We examine the parallels between digital immortality and each of these analogies in a proactive effort to prevent harms before the widespread public adoption of this technology. The contributions of this paper are a preliminary set of recommendations based on the limitations and challenges presented by a related set of real-world and sociotechnical systems involving social relationships, death, and grief.