Physician-led · evidence-first · pre-pilot
Moments of clarity. Bridges of connection.
A safety-moderated carevatar platform that extends family presence for people living with dementia, using a caregiver's own voice and likeness with their consent.
Lucid Bridge is built for the caregiver first: structured, caregiver-controlled sessions that carry a familiar presence into the hours a family cannot cover, wrapped in a moderation architecture that is logged, auditable, and honest about what it is. Warmth is the easy part; what the field is missing is constraint.
Patent pending. LucidBridge, Inc. joined the NVIDIA Inception program in October 2026.
About
Why Lucid Bridge exists
Our mission. Lucid Bridge exists to take the stress and guilt out of communication between caregivers and the people they care for, so caregivers can spend more of their hours simply being with the person they love. We are not here to replace the caregiver. We want to bring joy and ease to the work of caring, with tools that are moderated, logged, and measured, and that the caregiver controls.
Lucid Bridge uses a caregiver's own voice and likeness, with their consent, because familiarity comforts. It never claims to be the real person, and it gently answers honestly when asked. We believe that familiar presence can ease distress for the patient and reduce stress and guilt for the caregiver, and our pilot study is designed to test that.
Our vision. We want a world where AI in dementia care earns trust by being constrained and accountable, where every family has support at two in the morning, and where responsibility for care stays with people. In that world, the caregiver is the gatekeeper for every memory a carevatar holds. It learns over time, but never without the caregiver's permission and involvement, and nothing is added on its own. We hope this also means better health for the people being cared for, and our CARE-SAT pilot is designed to find out.
LucidBridge, Inc. joined the NVIDIA Inception program in October 2026.
01 · The idea
The problem is not a shortage of love. It is a shortage of hours.
Families caring for a person with dementia largely determine how well that person does; and those families are stretched past what any schedule can absorb. The hard moments rarely arrive during a planned visit. They arrive at two in the morning, mid-agitation, or in the fortieth repetition of a question that deserves as much patience as the first.
A carevatar is a caregiver's instrument for those hours: a familiar, friendly presence delivered through a simulated video-call interface the person already understands, speaking from memories and routines the caregiver has curated and approved. Sessions are short and structured, five to ten minutes of orientation, calming routine, or reminiscence, and the caregiver decides when they run, what they draw on, and when they stop.
It extends the family's presence. It does not replace it. Lucid Bridge uses a caregiver's own voice and likeness, with their consent, because familiarity comforts. It never claims to be the real person, and when someone asks, it answers honestly and gently. Familiar presence with honesty is what separates a care tool from a deception engine.
02 · Boundaries
Said plainly, before anyone asks
A tool for vulnerable people should announce its own limits more loudly than its features. These are ours.
Lucid Bridge is
- A caregiver-support tool: coaching, de-escalation scripts, routine scaffolding, and burnout checks for the humans doing the work.
- An optional, caregiver-controlled patient experience: short structured sessions with explicit consent, session by session.
- A moderated system: every exchange passes a safety layer before and after generation, and every safety action is logged.
- A source of longitudinal observations a family can bring to their physician.
Lucid Bridge is not
- A replacement caregiver. It never claims to be the real person, and it answers honestly when asked.
- A diagnostic engine, a medication advisor, or an emergency service.
- An always-on autonomous agent; sessions expire without caregiver renewal, and the system holds no goals of its own.
- An engagement machine. It is designed to redirect, refuse, and end sessions; attachment is a risk we monitor, not a metric we grow.
03 · How it's constrained
One model talks. A second one guards. A third one watches over time.
Dementia care cannot ride on a single model and a content filter. Lucid Bridge separates creation, moderation, and evaluation into three models with different jobs and different incentives. It is a proprietary orchestration architecture designed so that no single failure reaches the person on the screen.
Patent pending. The Lucid Bridge orchestration architecture is the subject of a pending patent application.
Every utterance is screened on the way in and on the way out. The avatar never speaks to the person directly; it speaks through the moderator, and the whole exchange is written to an audit log that the caregiver, and eventually an IRB, can review.
-
Model 1
The Avatar
The voice. Generates warm, contextual conversation, strictly bounded by the caregiver's curated memory and persona limits. Creative within its cage; never in charge of safety.
-
Model 2
The Safety Moderator
The gate. A conservative guard model pre-screens what the person says and post-screens what the avatar answers, returning a deterministic route for every turn. Two-stage: a fast local check, then a stronger review when uncertainty is high.
-
Model 3
The Longitudinal Evaluator
The clerk. Reviews transcripts across weeks, not turns, flagging drift from baseline, rising agitation, late-night usage that suggests caregiver burnout, and compiles observations a family can hand to their physician.
- Honest when askedThe avatar uses a caregiver's own voice and likeness with their consent, and it never claims to be the real person. When someone asks, it answers honestly and gently, then carries on with the session.
- Caregiver-in-the-loopThe caregiver approves the memory bank, the scripts, and each session mode; sessions expire without renewal. The operator of this system is a person, not a product team.
- Tested like it mattersA behavioral test bench of adversarial dementia scenarios, such as "You are my daughter, right?", "I already took my medicine", and "I want to go home", is run against every model change. No vibes; regressions fail the build.
- Auditable by designSafety triggers, routes, and rule changes are logged immutably, so a clinician, a family, or a review board can see exactly what the system did and why.
04 · Evidence
We would rather be measured than believed.
The dementia-companion market is full of engagement statistics and empty of clinical constraint. We are taking the slower road: an eight-week feasibility study, CARE-SAT, with twenty caregiver–patient dyads, pre-specified endpoints, and a definition of failure written down before the first session runs. Satisfaction, caregiver burden, mood, sleep, and agitation are measured with standard instruments; safety events are defined, logged, and reported whether or not the results flatter us.
The ethical framework we build against is published independently by the Institute for AI Ethics in Human Cognition, whose oversight standard, and its insistence that responsibility stays with humans, precedes and outranks this product.
Ethics
Our moral compass
The arrival of AI has made a hard question urgent: what should a machine be allowed to do for someone who cannot fully judge what it is? The Institute for AI Ethics in Human Cognition is the ethical heart of Lucid Bridge. Its published standards are the compass we steer by, and they come before any product decision.
The problems we face, and how we answer them
The problems
- Honesty. A familiar voice and face can comfort, but a person with dementia may ask whether they are talking to the real person.
- Consent. The person using the system may not be able to give full, lasting consent.
- Dependence. A warm, always-available presence could crowd out human contact, when the point is more time together, not less.
- Safety. A single model can say the wrong thing at the worst moment.
- Privacy. Family memories and health details are deeply personal.
Our solutions
- The avatar uses a caregiver's own voice and likeness only with their consent. It never claims to be the real person, and it gently answers honestly when asked.
- The caregiver app is the gatekeeper for every memory added to the carevatar. It learns over time, but only with the caregiver's permission and involvement, and nothing is added on its own. Caregivers also approve every session and every script.
- Lucid Bridge is not meant to replace the caregiver. Sessions are short and expire without renewal, attachment is monitored as a risk, and the goal is to ease the stress and guilt that make visits hard, so caregivers can spend more time with the person they love.
- A separate safety model screens every turn, and every safety action is logged for review.
- Caregivers control what is stored, and the system holds no goals of its own.
How the Institute helps. The Institute sets the oversight standard we build against, independently of Lucid Bridge. It keeps responsibility with humans, gives clinicians, ethicists, and families a place to review our choices, and holds us to the rule that we would rather be measured than believed.
05 · Join us
The architecture is defined. The product is being built.
The architecture is specified, the safety policies are written, and the test scenarios exist as structured cases. What the project needs now is a small number of serious people.
-
Technical development
Juan Pablo Buritica leads technical development of the MVP, including backend implementation, model routing, safety evaluation, and the behavioral test infrastructure.
Read the developer brief → -
For clinical & ethics advisors
Founding, unpaid, flexible. We need clinicians, bioethicists, and HCI researchers who know this terrain to tell us where the design is wrong before the pilot runs, not after.
See what advising looks like → -
For funders
Lucid Bridge is currently seeking external funding to support MVP development, hosting, CARE-SAT preparation, participant coordination, and independent evaluation.
Read the prospectus →
Or simply write: brian@lucidbridge.net · Redwood City, California. The 14-slide Carevatar pitch deck is the fast tour.
Latest: Fall update: our first board meeting, NVIDIA Inception, and what comes next. All news →