NeuralSight AI investor deck — print edition
A child sat in that classroom for seven months
before anyone noticed anything at all.
Scroll
Abu Dhabi · Sharjah · United Arab Emirates
NeuralSight AI
The world's first Agentic AI infrastructure for child safety and behavioral insights. Before psychological pressure, behavioral shifts or external crises turn into trauma — we see it, we understand it, and we guide early intervention.
Children are fighting battles the world knows nothing about.
Disruptive.
Distracted.
Difficult.
Trauma at home.
Abuse outside school.
An undiagnosed neurological profile.
The window where a quiet word
changes a childhood —
passed unnoticed.
Schools run on random human observation.
Traditional security cameras catch only blatant physical violence. Paper and digital surveys catch only what a child is willing to write down. Parents discover the truth years after the trauma has already taken root.
The signal was always there.
Nobody was built to read it.
No competitor operates at this depth.
Passive recording. Spot only blatant physical violence, after the fact.
Paper and digital questionnaires. Depend entirely on a child having the words and the courage.
Anonymized computer vision plus an Agentic AI infrastructure monitoring baseline behavioral shifts, trauma, safeguarding risks and neurodivergence — live, inside the classroom.
As monitoring hours accumulate, the network grows exponentially smarter.
An integrated network of autonomous agents.
Trauma & Baseline
Detects when an active child suddenly turns withdrawn and broken.
- Baseline shifts against the child's own behavioral history
- Total isolation periods and self-cowering
- Avoidance of lifting the head during lessons
- Indicators of depression and family stress
Safeguarding & Abuse
Reads the body language a child cannot put into words.
- Startle responses to sudden approach
- Touch aversion and immediate limb retraction
- Sustained non-verbal distress markers
- Confidential routing to the child-protection officer
Bullying & Aggression
Maps aggression as a spatial-motor pattern, not an incident report.
- Repetitive peer-group steering into the same dead zone
- Blind-zone mapping across the campus
- Progressively shrinking defensive posture
- Timestamped timeline for the counselor
Neurodivergence · ADHD & Autism
Turns years of guesswork into a clinical evidence log.
- Prolonged zoning-out periods during lectures
- Seat departures correlated with motor energy spikes
- Repetitive self-calming movement (stimming)
- Three-month objective evidence log for specialists
Absolute privacy, by architecture.
Student bodies are instantly converted into anonymized skeleton vectors.
UAE data protection law, GDPR in Europe, FERPA and COPPA in the US.
Runs on the school's existing camera network via Edge AI processing units — no cloud streaming, no new cameras.
NEVER STORED
0 BYTES OF IMAGERY
From an existing camera to a clinical-grade report.
Edge integration
Existing CCTV. No new hardware.
A compact Edge AI Box installs on-site and connects to the school's current indoor camera network in classrooms and corridors. Zero new CAPEX, zero cloud-streaming cost.
Instant anonymization
Privacy before the frame moves.
Bodies are converted into anonymized skeleton vectors on the box itself. No raw footage is ever stored and no facial recognition is used — GDPR, FERPA and UAE PDPL by architecture.
Baseline & monitoring
Two weeks to learn one child.
The first two weeks of term build a behavioral baseline per student: movement, interaction rate, energy level. The agent network then watches continuously for deviations.
Structured reporting
Data becomes intervention.
Anomalies are translated into specialized reports routed to administration for on-site action, parents through the mobile app, and clinical specialists as standardized evidence logs.
What the school never sees, the system does.
What the school sees
A quiet week. No incidents reported.
What is actually happening
Never hit in front of a teacher. Ostracized, herded into courtyard blind spots, belongings snatched — and far too afraid of retaliation to tell anyone.
How the system detects it
- Repetitive spatial-motor patterns: the same peer group steering one student into the same dead zone across multiple days
- Blind-zone mapping of where the encounters happen
- Defensive body language and progressively shrinking posture against the student's own baseline
Instant & long-term solution
- Confidential alert to the administrative supervisor with exact location, zone number and the students involved
- Timestamped bullying timeline report so the counselor can break the dynamic and rehabilitate the class environment
What the school sees
A polite child. Good attendance. No complaints.
What is actually happening
Harm suffered outside school, or from a close circle. The child lives in silent terror, has no vocabulary for it, and says nothing at all.
How the system detects it
- Startle responses: violent motor flinching when someone approaches suddenly
- Touch aversion: physical cowering and immediate limb retraction
- Sustained non-verbal distress markers against baseline
Instant & long-term solution
- High-priority safeguarding alert sent exclusively to the child-protection officer and senior psychologist
- Movement and defense-pattern breakdown to support a gentle, confidential parent consultation and specialist referral
Objective evidence instead of years of guesswork.
What the school sees
“Unfocused and hyperactive.” A discipline problem.
What is actually happening
Reprimanded at school for being unfocused and hyperactive, argued with at home over unfinished homework. Nobody sees the neurological framework underneath.
How the system detects it
- Prolonged zoning-out periods during lectures
- Frequent seat departures correlated with unfocused motor energy spikes
- Repetitive self-calming movement (stimming)
Instant & long-term solution
- A Clinical Evidence Log documenting behavioral and motor patterns across the past three months
- Securely shared with parents alongside a specialist referral — years of guesswork replaced by objective data
What the school sees
A high performer who “finally settled down.”
What is actually happening
A high-performing, energetic student hit by a crisis at home. Within weeks: quiet, isolated, head buried in books, no participation at all.
How the system detects it
- A sharp 70% drop in energy and movement metrics over a two-week window
- Emerging preference for classroom corners
- Complete social withdrawal versus historical baseline
Instant & long-term solution
- Early-warning report to student guidance flagging a radical, sudden drop in psychological and behavioral metrics
- A supportive parent session, a customized in-school support plan and temporarily eased academic workload
Three surfaces, one intelligence layer.
Risk dashboard
Identifiers are pseudonymous. No imagery, no faces, no names in the pipeline.
Skeleton timeline
- Posture contraction
- +38% vs baseline
- Blind-zone dwell time
- 4 days recurring
- Peer proximity cluster
- 3 identical students
Weekly wellbeing
A category with no incumbent.
Early years, where intervention still changes everything — 450M+ students addressable worldwide.
From first pilot to sovereign scale.
MVP & First Pilots
Edge AI box and agent network shipped to first UAE pilot campuses.
MENA Launchpad
From the UAE into Saudi Arabia, Qatar and Kuwait.
GDPR Scale
Expansion into Europe and the United Kingdom.
Global Sovereign Deployment
Scaling across the Americas and Asia.
Four revenue streams, one deployment.
Psychological climate report and medical guidance. CAC is zero — driven by official school circulars.
Classroom psychological climate mapping and early alerts for intervention before incidents occur.
Verified environmental diagnostic reports requested by parents to support clinical sessions.
Central national radar for ministries of education to track quality of life and safety indices.
Move the sliders.
Modelled on the stated pricing: $30/month per parent subscription and an average $32,000 annual school licence at ~600 students per school.
Edge AI economics.
Forward projections, not history. Year 1 begins the moment the MVP goes live on the first pilot campuses.
| Metric | Year 1 | Year 2 | Year 3 |
|---|---|---|---|
| Covered students | 6,000 | 45,000 | 220,000 |
| ARR | $0.25M | $2.1M | $11.5M |
| Gross margin | 86% | 90% | 93% |
| OpEx | $620,000 | $1.4M | $3.6M |
| EBITDA | -$405,000 | $490,000 | $7.1M |
| Net profit margin | — | 23% | 62% |
Break-even in Year 2; 62% net margin by Year 3.
$1.2M pre-seed.
18 months of runway: MVP on live campus cameras, first paid pilots, seed readiness.
- /A working MVP running on live campus cameras
- /First signed pilot schools across the UAE
- /Validated behavioral baseline accuracy
- /Seed-ready traction and reference deployments
“NeuralSight AI is not just a startup. It is the intelligent infrastructure that turns every silent camera in our schools into a watchful eye, an understanding mind, and a shield protecting our children before it is too late.”