Home/Technology/AI Semiconductors

Silicon designed for disaster.

Roze AI is developing a disaster-prevention SoC for its DAP platform. The initial RF-and-MCU implementation was completed in 2023 and underwent proof-of-concept validation and internal testing. The program is currently in the engineering-sample stage.

Energy-harvesting research

Energy-harvesting disaster-care SoC research.

Energy harvesting is part of Roze AI’s research and development roadmap. The company is developing technologies intended to support lower-power, longer-duration sensor operation. Commercial availability and performance remain subject to development and validation.

The research program is intended to support lower-power, longer-duration sensor operation. Commercial availability, maintenance requirements and performance remain subject to development and validation.

  • Roze AI is researching energy-harvesting technologies for low-power sensor applications. The specific energy sources and implementation remain subject to development and verification.
  • Initial implementation and PoC/internal testing disclosed; commercial deployment requires confirmation
  • Development is intended for Roze AI sensor applications; manufacturing is expected to be outsourced.
Roadmap

Three generations to a dedicated disaster-AI chip.

Development roadmap · targets subject to change

First-version AI ASIC / SoC

Target launch: Q1 2028. Development milestones do not represent commercial availability.

Development roadmap · targets subject to change

Energy Harvest Chip

Target launch: Q4 2029. Development milestones do not represent commercial availability.

Future roadmap

Disaster-AI processing

Future semiconductor generations may add on-device AI processing. Timing and specifications remain subject to the disclosed roadmap.

SoC development program™

The on-device protector chip.

SoC development program is described as an edge-processing direction for fire and disaster data. Actual latency, connectivity resilience, and product availability require product-level testing and evidence.

  • Planned on-device analysis of multi-sensor telemetry, subject to technical validation
  • Early anomaly detection independent of cloud availability
  • Feeds the Fire4Cast™ platform when connected; protects alone when not
Back to platform overview →