Life-size digital expert docents for physical venues
When CheshireCat® recognition is present, Alice® Avatars use visitor context to personalize each interaction. Combined with CheshireCat® recognition technology, avatar docents know who they're talking to. A returning visitor gets continuity: "Welcome back - last time we stopped at the lunar module. Ready to continue?" A family group gets age-appropriate content for the kids while delivering expert knowledge to the parents. A visitor who switches expert mid-visit carries their history: the new guide knows what has already been covered.
Recognition context is processed on-premises. No biometric data is retained beyond the session. No cloud dependency. CheshireCat® operates with privacy as the default - the venue receives the answer ("returning visitor, interested in engineering"), never the document.
Alice® Avatars include full ASL (American Sign Language) signing capability. Deaf and hard-of-hearing visitors receive the same rich, personalized storytelling as every other guest - delivered by their chosen expert docent in fluent sign language. This is not a subtitle overlay or a separate accessibility mode. The avatar itself signs, with proper facial grammar, co-articulation, and natural motion.
Walk into Mad Systems HQ and you meet Samtoo, our Virtual Receptionist - a live Alice® Avatar. Samtoo greets staff and visitors by name, checks schedules, and adapts conversation to each person - a working demonstration of Alice® Avatar capabilities in daily use. Need Korean for tomorrow's delegation? The language pack was uploaded during lunch, with no downtime and no vendor service call.
Multiple expert personas - each drawing from the same venue-controlled Alice® Body of Knowledge - available from a single installation. Each visit can engage a different depth level, interest track, or narrative thread. Available to every visitor, at any time, without limiting access by group size or scheduling.
Full editorial control. Update the knowledge base overnight. Spin up limited-time avatars for events. Analytics show which objects generate the most engagement and where visitor attention shifts.
Long-lifecycle, non-proprietary compute infrastructure. Modular upgrades. Revenue upside through repeat visits, premium ticket tiers, and brand-sponsored personas.
ASL signing avatars, multi-language support (100+ languages), Bluetooth hearing aid streaming, and age-appropriate content delivery - all from one system.
There is a well-known parable about five monkeys. It is not really about monkeys. It is about what happens when a group keeps enforcing a rule long after the original reason for that rule has disappeared. In AV and Location Based Entertainment, that pattern shows up all the time.
Five monkeys are placed in a habitat with a ladder in the middle. At the top of the ladder hangs a bunch of bananas. Naturally, one of the monkeys climbs the ladder to grab the bananas. When it does, the researchers spray the rest of the monkeys with cold water. After a few attempts, the monkeys quickly learn the connection.
Before long, whenever a monkey even tries to climb the ladder, the other monkeys pull it down and beat it up to stop it.
But none of them will climb the ladder. And if a new monkey tries, the group will still beat it up.
That is how large parts of this industry still operate.
Too many teams inherit technical assumptions from the last project, then production assumptions from the last technical stack, then budget assumptions from the last production model. Eventually those assumptions stop looking like choices and start being treated like physics.
We chose a different starting point.
We built a system that bases AV hardware on non-proprietary compute node hardware rather than on closed black box endpoints. That decision changed the problem space.
We did not set o