“Was the system compromised?”
- Unauthorised access
- Technical anomaly
- System integrity
- Isolated events
Behavioural risk intelligence
Cybersecurity sees access.
We look for harm.
Treonix is building Harm Grid—a person-centred intelligence platform designed to surface escalating patterns of technology-enabled harm occurring through legitimate or authorised access.
RISK INTENSITY
TIME / BEHAVIOURAL PROGRESSION
ACUTE DEVIATION
Pattern emergingIllustrative trajectory—not an allegation
01 / THE GAP
Not every digital threat begins with a breach.
Conventional cybersecurity is optimised to detect unauthorised access, malware and compromise. But technology can become a mechanism of coercion while functioning exactly as designed.
The innovation gap is not simply technical. It is perceptual.
SHIFT
THE
LENS →
THE
LENS →
“Could legitimate functionality be producing harm?”
- Authorised-access misuse
- Behavioural progression
- Human safety
- Patterns over time
02 / THE PLATFORM
From events to
harm intelligence.
Harm Grid translates fragmented signals into an explainable temporal picture—without replacing the people responsible for interpretation and action.
Harm Grid
LAYER 05
Harm Grid
A temporal view of trajectory and progression.
03 / DESIGN PRINCIPLES
Human safety demands
a different architecture.
Person-centred
The unit of concern is potential human harm—not simply the integrity of a device.
Pattern-aware
Isolated events become meaningful when read as persistence, escalation and trajectory.
Explainable
Risk signals remain interpretable, contestable and subject to human judgement.
Ethically bounded
Harm Grid identifies patterns associated with harm. It does not label a person an abuser.
04 / RESEARCH FOUNDATION
Research translated
into a viable technology.
The Harm Grid grows from original work spanning survivor-centric threat modelling, coercive-control classification, fuzzy risk scoring and acute deviation.
RESEARCHThreat models that recognise authorised-access harm
→THEORYCCKC + FRS + Acute Deviation
→PRODUCTExplainable temporal harm intelligence
Our boundary is deliberate.
Treonix detects patterns associated with potential harm. It does not determine intent, guilt or identity. Human judgement, survivor safety, consent and contestability remain central to every deployment.
Research to responsible innovation
Help build a safer
digital future.
We are beginning conversations with researchers, advocates, technologists and responsible partners.
Talk with Treonix ↗