The digital legacy space moves fast and the vocabulary hasn’t caught up. Some terms get used interchangeably when they shouldn’t. Others sound familiar but mean something specific in this context. This glossary defines the terms that matter — clearly, without the marketing layer.

AI Avatar

A visual representation of a person, generated by AI, that can move, speak, and respond in real time. AI avatars in digital legacy are distinct from static photos or pre-recorded videos — they are interactive, meaning a family member can ask a question and receive a response that reflects the person’s face, voice, and expression together.

The difference between a well-built avatar and a poorly built one is not always obvious from a demo. The question worth asking is not “does it look like them?” but “does it sound like them when asked something they never specifically answered?”

AI Memorial

A digital space — typically interactive — created in memory of someone who has passed. AI memorials range from simple chatbots trained on text messages and social media posts to full personality simulations with voice and video. The term covers a wide spectrum of quality and architectural approach.

Most AI memorials are built retrospectively, after a person has died, from whatever data is available. The limitation of that approach is real: the system is working from what was left behind, not from what was intentionally built. The difference between a memorial and a living legacy is worth understanding clearly.

The framework a platform uses to determine who controls what gets preserved, who can access it, and when. Consent in digital legacy is not a checkbox. It is an architectural decision that determines whether the person being preserved — or someone else entirely — has genuine authority over their own representation.

Platforms differ significantly here. Some give custodians broad access from day one. Others require specific triggers. Some allow the person to set precise conditions for access, including time delays, inactivity windows, and named individuals only. This is one of the most important questions to ask before choosing a platform, and one of the least visible from the outside.

Digital Afterlife

Originally, the term referred to the collection of accounts, photos, emails, and files a person leaves behind when they die. In 2026, it has expanded to include AI-generated representations that can speak, respond, and interact — a shift from passive archive to active presence.

The distinction matters practically. Managing someone’s digital afterlife used to mean closing accounts and downloading photos. Now it can include deciding whether an AI version of that person continues to exist, who can talk to it, and on whose terms. What that actually looks like in practice, and what families are navigating now, is a longer conversation.

Digital Estate

The full collection of a person’s digital assets — accounts, files, subscriptions, cryptocurrency, domain registrations, and any AI representations built in their likeness. A digital estate is increasingly treated as a legal matter alongside a physical estate, with specific laws in many jurisdictions governing who can access what and under what conditions.

Most people have not planned their digital estate. The accounts keep existing after death. The content stays up. The passwords go with the person.

Digital Executor

A person designated to manage your digital assets after death. The role is distinct from a traditional executor: a digital executor handles online accounts, cloud storage, AI representations, and digital files rather than physical property. Some jurisdictions formally recognize this role. Others do not yet, which creates legal gray areas families are navigating regularly.

Choosing a digital executor is a practical step. Communicating clearly with them about what you want preserved, deleted, and shared is equally important — and harder to do after the fact.

Digital Immortality

The broader concept of preserving a person’s identity, memories, and personality indefinitely through technology. Digital immortality is not a single product or platform — it is an idea that spans a range of technical approaches, from simple recording archives to AI personality simulation.

The term is used loosely in the industry. What one platform calls digital immortality, another calls a legacy bot. What matters is not the label but what the system can actually do: whether it responds or just retrieves, whether it stays consistent over time, and whether the person built it themselves or it was assembled from whatever data survived them.

Digital Remains

The data — accounts, messages, files, recordings, and AI representations — that persist after someone dies. Digital remains are increasingly a legal and ethical category, not just a practical one. Who owns them, who can access them, and whether an AI representation constitutes a form of digital remains with its own rights are questions regulators are beginning to address.

The EU AI Act and emerging frameworks in other jurisdictions are starting to define what can be done with a person’s digital remains without their prior consent — a conversation the industry has been slow to have.

Digital Will

A document that specifies what should happen to a person’s digital assets after death. A digital will may address which accounts to close, which to memorialize, who receives access to files, and whether any AI representation of the person should be created, maintained, or deactivated.

A digital will has no standard legal form, and in most jurisdictions it sits alongside rather than inside a traditional will. Its value is practical: it reduces the uncertainty family members face when trying to manage someone’s digital presence without clear instructions.

Ghost Bot

An AI simulation of a deceased person, typically assembled from their existing digital footprint — social media posts, messages, emails, recordings — without prior planning or consent from the person themselves. Ghost bots have grown rapidly as a category, powered by the same generative AI tools available to anyone with an internet connection.

The ethical problems with ghost bots are significant. The person never agreed to be represented this way. The simulation is only as accurate as the data that was scraped — which means it reflects what someone chose to post publicly, not who they actually were. For families who encounter them, the experience is often described as uncanny and disorienting rather than comforting. What distinguishes a deliberately built legacy from a ghost bot is worth understanding before the question becomes urgent.

Grief Tech

A category of technology products built around the experience of loss. Grief tech includes AI memorial services, posthumous communication tools, legacy archiving platforms, digital estate managers, and a growing range of companion tools designed for people who are actively grieving.

The category is broad and quality varies widely. Some grief tech products are built with genuine care for the people who use them. Others have been criticized for exploiting vulnerability, moving too fast, and providing a comforting illusion rather than something that lasts. The distinction is not always visible from the outside.

Identity Drift

The gradual degradation of an AI representation’s accuracy over time — when the system’s responses begin to reflect pattern and inference rather than the specific person it was built to represent. Identity drift is common in AI systems that simulate personality from surface traits without a structured memory foundation. It can be subtle at first but compounds over years.

A representation that feels accurate at launch is not the same as one that remains accurate a decade later. Identity drift is the problem the industry has been slowest to name clearly, and the one that matters most for a true digital legacy.

Legacy Bot

An AI chatbot trained on a specific person’s digital data — messages, emails, journals, recordings — and designed to respond as that person would. Legacy bots are among the earliest and most common forms of AI memorial. Their output is shaped entirely by the quality and completeness of the data they were trained on.

A legacy bot built from rich, intentionally curated input is a different product from one assembled after death from whatever is available. The term covers both. Asking which kind you are looking at is a reasonable question before trusting what it produces.

Legacy Contact

A person designated by a platform user to manage or access their account after death. The term originates with major platforms like Apple and Facebook, which allow users to nominate someone with limited, post-death access to their data. Legacy contacts on consumer platforms typically have restricted capabilities: they can download content or manage memorialization, but cannot access private messages or make substantive changes.

The concept in dedicated digital legacy platforms is similar but more expansive. A designated person can receive access to a full AI representation, not just account data. The terms of that access — who, when, under what conditions — vary significantly by platform.

Memorialization

The process of converting an active account or profile into a permanent tribute after someone’s death. Memorialization typically prevents new logins, disables account changes, and signals to visitors that the account belongs to someone who has passed. Major platforms offer this for social media accounts.

Memorialization and digital legacy are different things. A memorialized account preserves what was posted. A digital legacy, built deliberately and in advance, preserves who the person actually was.

Personality Preservation

The goal of capturing who someone was — their values, opinions, reasoning, humor, and specific beliefs — in a form that can respond accurately to questions that were never explicitly answered. Personality preservation is distinct from style imitation, which captures how someone sounds, and from archive retrieval, which replays what they said.

The gap between a system that imitates personality and one that preserves it is significant. The architectural choices that determine which side a platform lands on are worth understanding before making a decision.

Posthumous Interaction

A conversation between a living person and an AI representation of someone who has died. As the technology has matured, posthumous interaction has moved from a concept discussed in academic papers to a product category used by real families. The emotional weight of the experience depends almost entirely on how accurately the representation was built.

The people who report the most meaningful posthumous interactions are consistently those whose loved ones built their own legacy deliberately, over time, rather than those working with systems assembled after death from leftover data.

Sunset Clause

A provision specifying that an AI representation of a person should be deactivated after a defined period or condition — for example, once their children reach adulthood, or after a set number of years. Sunset clauses reflect a growing recognition that an eternal AI presence is not always what families want or what the person would have chosen.

Including a sunset clause in a digital legacy plan is a meaningful act of care. It acknowledges that a legacy is for the people who are grieving, and that what they need may change over time.

EchoVault is a digital legacy platform built to last — not just to impress on day one. Start building your Echo.