Skip to content
Technology How it works Breeder — Hyperion Burner — Aegis Burner — MetroVolt AI-Native Architecture Magnets Fuel cycle Safety Roadmap
Solutions AI & Data Centers Defense & Government Grid & Baseload Neutron Detection Quantum
Learn Technical Library
Proof Publications Whitepapers Technical Library Open Science & Reproducibility The Honest Gates
Company About / Mission Leadership Environment Health & Safety Investors Careers Press Contact
3D Model
AI Architecture › L6 · Experience
L6 · Experience

Attention and Notification Design

Using visual, auditory, and spatial cues in a disciplined budget so the machine can grab an operator's attention exactly when — and only when — it should.

THE STACK · click to jumpL7Ecosystem & StrategyL6Experience & VisualizationL5Applications & CopilotsL4OrchestrationL3Twin Modeling & AIL2Data FabricL1Control PlaneL0Foundation▲tlmctl▼L6 · EXPERIENCE & VISUALIZATIONHow people see, steer, and review the plant.1Control-Room 3D Twinlive overlays2Plant-Floor SCADAoperations HMI3Mobile Engineeringfield access4Alerting UXtriage & escalation5DashboardsKPIs & health6Replayincident reviewMACHINE TIESurfaces the L3 twin state and L5 copilots to human operators.KRONOS FUSION ENERGYAI-NATIVE S.M.A.R.T. GENERATOREXPERIENCE & VISUALIZATIONSHEET 08REV. 2026-08L6 · AI-NATIVE STACK
L6 · Experience & Visualization — its place in the stack (left, click any layer) and its internal components (right). Telemetry rises; control descends.

Attention is a scarce, protected resource

Every notification spends operator attention, and attention does not refill on demand. If the interface interrupts too often it trains the operator to tune it out (the alarm-fatigue failure mode); too rarely and it misses the moment to re-engage. Kronos L6 treats attention as a budget and spends it deliberately, matching the intrusiveness of a cue to the urgency and confidence of what triggered it.

A graded cue hierarchy

UrgencyCue
Ambient / informationalpassive change on the shared picture, no interrupt
Notableperipheral highlight; enters the triage stream
Action needed soondistinct tone + persistent card with countdown
Immediate actionstrong multi-modal cue reserved for this level only

The strongest cues are reserved and rare, so that when they fire operators respond reflexively instead of second-guessing. Auditory cues are distinct and few — a small vocabulary of sounds an operator can identify without looking — because a room full of similar beeps conveys nothing. Spatial cues matter too: an urgent item is drawn where the responsible operator is already looking, not stranded on a peripheral screen.

Confidence modulates intrusiveness

A high-confidence, high-consequence detection earns a strong interrupt; a low-confidence one is surfaced without hijacking the operator, because a strong interrupt on a shaky detection is how trust erodes. This links notification strength directly to the model's own uncertainty and provenance — the same signals shown in uncertainty visualization. When a model is out of distribution, its notifications are marked as such so the operator weights them appropriately.

The notification budget is validated in the simulator against scripted breeder disruptions and burner excursions, tuning cue thresholds so operators are re-engaged before events without being worn down between them. See cognitive load and flood suppression.

Content reviewed August 2026 · design-and-simulation stage