A science-fiction story in development

Real components. A fictional whole.

The science makes the questions sharper. It does not certify a Luminai.

Confirmed story information, developing ideas, and unanswered questions are labeled separately.

What the evidence can support

Human brain–computer interfaces have decoded trained attempted actions. A 2023 speech study reported 62 words per minute in one participant, with substantial errors on a large vocabulary. It did not read unrestricted private thought. Primary paper.

A brain–spine interface supported standing and walking in one person with chronic spinal cord injury. That is a functional neural bridge, not general cognitive union. Primary paper.

Nine months of tactile Braille training in sighted adults was associated with anatomical changes. It supports learning-related plasticity, not direct reception of arbitrary wireless information. Primary paper.

Keep demonstration types separate

Evidence tierExampleBoundary
Human demonstrationTrained speech decoding; brain–spine movement support.Specific tasks and limited participants.
Animal experimentUltrasonic neural dust recorded nerve and muscle signals in anesthetized rats.Peripheral recording, not human cognitive bonding.
Mixed preclinical and human feasibilityA 1,024-channel cortical-array study reports porcine/cadaver delivery and a five-patient intraoperative pilot.Does not establish a chronic, invisible, non-surgical everyday interface.
Engineering prototypeRoom-scale wireless power in an engineered resonator.Does not show that an entire city safely powers microscopic brain nodes.
Fictional synthesisA personal bond integrated with environmental computation and sensing.Components have not been demonstrated together as a Luminai.

Primary sources: neural dust, cortical array, room-scale resonator.

Open architecture questions

An interface needs physical transduction before the brain can learn its signal. Power transfer, data transfer, sensing, and neural stimulation are different jobs. A single unspecified field cannot answer all four.

Candidate systems face range, alignment, coupling, heating, interference, signal drift, and maintenance constraints. Moon-based oversight also needs local safeguards for delays and disconnection. These are design questions, not approved failure modes.

Claims deliberately left unproven

The brief is preliminary. Its complete source list has not become a verified catalogue merely by being saved. The evidence audit identifies verified claims and references still needing checks.

Neither a biological identity check nor individualized learning guarantees security. Neither intelligence nor deeper integration guarantees virtue. The story has to earn those outcomes through conduct and accountable institutions.

Source: 48 - Luminai Evidence Audit and Architecture Boundaries. Explore the architecture workshop.