Afterlife AI(TM) vs Meta's Afterlife Patent: Consent, Not Reconstruction

Meta's afterlife patent describes reconstructing a person from their posts, messages, and likes to simulate them after death, with no consent step from the subject. Afterlife AI(TM) is the inverse: a Persona you build yourself while alive, governed by Executor Lock(TM), authored with explicit consent.

The one-line difference

A griefbot reconstructed from your data is something done to you. A Persona you author while alive is something you choose. Meta's patent sits on the first side of that line. Afterlife AI(TM) is built entirely on the second.

In December 2025 Meta was granted a patent, filed in 2023 and credited to CTO Andrew Bosworth, describing a system that trains a language model on a user's posts, comments, chats, voice messages, and likes, then deploys it to respond on their behalf during a long absence or after death. Meta told reporters it has no plans to ship it. The patent is still the clearest public articulation of the reconstruction model, and it is the clean foil for how Afterlife AI(TM) works.

Consent vs reconstruction

The patent's design starts from data a person already left behind. It infers a personality from a digital footprint that was never created for this purpose, and the subject is not asked. Afterlife AI(TM) starts from the person. You sit down, while alive, and author your Persona deliberately across the eleven dimensions that make you you. Nothing is scraped, inferred, or assembled after the fact.

• Reconstruction model: trained on existing posts, messages, and likes. No authoring step. No consent from the subject. Built to fill an absence, including death.

• Afterlife AI(TM): built by the living person, with explicit consent that covers posthumous use, and locked from change at death by Executor Lock(TM).

Executor Lock(TM): governance the patent does not have

Reconstruction has no transition of authority. Whoever holds the data holds the persona. Afterlife AI(TM) is governed by Executor Lock(TM), a cryptographic authority-transition that locks at death. The Persona you authored is the Persona that remains. It cannot be re-trained, re-pointed, or quietly edited after you are gone, because the authority to change it was sealed by you, on your terms, in advance.

Voice: preserved by you, not synthesised from scraps

The patent contemplates simulated audio and video drawn from a person's recordings. Afterlife AI(TM) takes the consent-first path here too: voice is a consent-based preservation of yourself, recorded while you are alive and locked at Executor Lock(TM). Creating your voice is free for everyone. Listening is the paid experience, from Legacy at $14.99/mo, and the family inherits the time you have paid for. We never name the engine, and we never claim synthesis happens in Australia.

What it costs to build yours

You start free. The free tier is a one-time build budget, not a trial and not a monthly allowance: 60 memories, 100 conversations, 1 Trusted Contact, and Executor Lock(TM) setup. No card. Your free build never expires.

Public pricing is three tiers: Free, Legacy at $14.99/mo, and Eternal at $29.99/mo. Family inherits the time you have paid for. IDY(TM) is an Australian company, Australian-hosted, and your voice is treated as sensitive information under Australian privacy law.

Which side of the line do you want to be on

If a digital you is going to exist, the only question that matters is who authored it. Build it yourself, with consent, governed by Executor Lock(TM). Build Once. Live Twice.(TM)

Start your free build now at https://app.afterlife.ai.

Frequently asked questions

What is Meta's afterlife patent?

A patent granted to Meta in December 2025, filed in 2023 and credited to CTO Andrew Bosworth, describing a system that trains AI on a user's posts, comments, chats, voice messages, and likes to simulate them during a long absence or after death. Meta has said it has no plans to build it.

How is Afterlife AI different from the Meta patent?

The patent reconstructs a person from data they already left behind, with no consent step. Afterlife AI(TM) is authored by the living person with explicit consent that covers posthumous use, then locked at death by Executor Lock(TM). One is reconstruction, the other is consent.

Does Afterlife AI scrape my social media?

No. Nothing is scraped or inferred. You build your Persona yourself, deliberately, across the eleven dimensions, while you are alive. The free build budget is 60 memories and 100 conversations, no card required, and it never expires.

What is Executor Lock and why does it matter here?

Executor Lock(TM) is a cryptographic authority-transition that locks at death. It means the Persona you authored cannot be re-trained, re-pointed, or edited afterward. Reconstruction-based systems have no equivalent: whoever holds the data controls the persona.

What does it cost?

Public pricing is three tiers: Free, Legacy at $14.99/mo, and Eternal at $29.99/mo. The free tier is a one-time build budget of 60 memories, 100 conversations, 1 Trusted Contact, and Executor Lock(TM) setup. Family inherits the time you have paid for.

How does voice work?

Voice is a consent-based preservation of yourself, recorded while alive and locked at Executor Lock(TM). Creating your voice is free for everyone. Listening is the paid experience, from Legacy at $14.99/mo, and the family inherits paid time. We never name the voice engine.