Predictive fire evacuations are now a reality. The National Institute of Standards and Technology (NIST) has developed an AI model that identifies safe evacuation routes during a blaze. Called “Safe Step,” the model works with electronic displays to assess building conditions and redirect occupants to their safest evacuation exit, all in real-time.
Described in the Journal of Building Engineering, the model uses “reinforcement learning,” a type of AI that determines optimal routes through trial and error. It draws on building layout data and output from a NIST fire simulation tool to anticipate how a fire will develop and spread over time. Currently, Safe Step can identify safe evacuation routes in single-story floor plans, with a multilevel version in the works.
Previous approaches relied on traditional algorithms to find the shortest safe evacuation path and did not account for cumulative hazards that evacuees may encounter along a route. The new model, by contrast, forecasts how a fire will evolve and updates emergency exit displays accordingly.
“We asked ourselves, ‘Can we build a better algorithm that predicts how the fire evolves, and in a way that helps save more lives?’” said NIST mechanical engineer Wai Cheong Tam.
Standards: The Quiet Infrastructure of Safety
Breakthroughs such as Safe Step are both a sign of the times and a reminder of a vital safety challenge: as AI comes to the forefront of decision-making roles in safety-critical environments, the standards community must keep pace with frameworks to guide its responsible deployment.
International standardization work to support this includes ISO/IEC TR 5469, Artificial intelligence — Functional safety and AI systems, and ISO/IEC 42001, AI management systems — both developed by the International Organization for Standardization /International Electrotechnical Commission (ISO/IEC) Joint Technical Committee (JTC) 1, Information technology, Subcommittee (SC) 42, Artificial intelligence, where INCITS serves as the U.S. Technical Advisory Group to JTC 1, and ANSI holds the Secretariat. The work is ongoing: JTC 1 SC 42 is in the process of developing a series on AI functional safety.
ANSI-accredited standards developers are also contributing across various AI domains, including the IEEE 7000 series, which addresses ethical considerations in the design of autonomous and intelligent systems.
On the fire and life safety side, the physical infrastructure that AI models like Safe Step depend on is standards-supported. UL 924, Standard for Emergency Lighting and Power Equipment, has extra investigatory steps beyond general lighting certifications, qualifying equipment to the requirements of the NFPA 101 Life Safety Code, NFPA 1 Fire Code, National Electrical Code® NFPA 70®, Article 700 Emergency Systems, and the International Code Council (ICC) International Fire and Building Codes, including the International Building Code (IBC) and International Fire Code (IFC).
The NIST AI Risk Management Framework offers complementary voluntary guidance for managing AI risks across the lifecycle.
Together, these standards support the broader ecosystem that makes AI-driven safety tools deployable—ultimately, with safer outcomes during fire emergencies.