AI can automate fire department NERIS reporting by pulling structured data from CAD, RMS, and apparatus systems to pre-populate required fields, then drafting incident narratives from those data points, reducing the average company officer report completion time from 20 to 40 minutes to under 10. NFIRS was retired on February 1, 2026, and NERIS, the National Emergency Response Information System, is now the only national fire incident reporting system. Version 1.5 took effect October 1, 2026, adding optional new incident types, actions, and location fields; every change is additive, so existing integrations keep working. Every 2026 incident is reported in NERIS, so departments that have not yet built efficient reporting workflows are already feeling the added burden on company officers. FlorianAI supports NERIS data integration as part of its core operations capability.
What NERIS Is and Why It Matters
NERIS, the National Emergency Response Information System, is the USFA's replacement for the National Fire Incident Reporting System (NFIRS), which collected fire incident data from the 1970s until it was officially retired on February 1, 2026. NERIS is no longer a future transition. It is the only system the USFA accepts for national fire incident reporting, and every 2026 incident is recorded there.
NERIS is a meaningful improvement over NFIRS in several ways: it uses a more modern data structure, it is designed for interoperability with CAD and RMS systems, and it captures a broader range of incident types and outcome data. The USFA’s goal is a national incident database that actually reflects what fire departments are doing, meaning not just fires but the full range of emergency and non-emergency responses that modern departments handle.
For fire departments, the switch is significant work. NFIRS had its own complexity, but most departments had built workflows, however imperfect, around it over decades. NERIS uses new data fields, new narrative standards, and in many cases new or updated RMS software. Officers are already hitting friction in the field, such as finding that an action like extrication is not cleanly selectable for a motor-vehicle accident. For company officers who are already responsible for completing a report after every call, often at the end of a 24-hour shift, a new system with new requirements is a real operational problem, not a future one.
What Changed in NERIS Version 1.5 on October 1, 2026
NERIS Version 1.5 took effect on October 1, 2026, the first major schema revision since NERIS became the national standard. Every change is additive. No fields were removed, the API structure did not change, and existing RMS integrations keep submitting exactly as they did before. There is no deadline and no required action. The new fields are optional, and a department adopts them when its RMS supports them.
What Version 1.5 adds is granularity the fire service asked for. Broad categories were subdivided so the data reflects what crews actually ran. Water rescue now splits into ice, flood, and surf or rip current. Micromobility collisions involving e-bikes and scooters separate from motor vehicle collisions, and human-powered vehicle collisions involving bicycles, wheelchairs, and skateboards split out again. Agricultural vehicle fires separate from standard vehicle fires.
The actions and tactics list grew to cover work departments perform routinely and could never record as structured data. Non-patient EMS support was the single most requested addition in the revision. Vehicle, structure, and earth or trench stabilization are now selectable. So is rapid intervention team establishment, with a timestamp. Ground and aerial ladder use, which previously lived only in narratives, became a recorded action.
Chief Chris Wolfe of Greater Naples Fire Rescue District named the gap plainly in the USFA announcement: "Structure stabilization has never shown up in the data, even though fire departments perform this service routinely."
Location classification expanded as well, adding occupancy types for data centers, mega warehouses over 200,000 square feet, automated storage and retrieval systems, commercial greenhouses, aquatic facilities, and fitness centers, plus modifiers that flag short-term rentals, hoarding conditions, and excessive fuel loads. An in-service flag also replaces the delete-and-recreate workaround departments used for a station closure or a seasonal move, and NERIS IDs now persist through a status change.
These fields populate going forward, from incidents recorded after a department's RMS adopts them, so the Version 1.5 dataset builds over months. Full technical guidance is in the USFA announcement and the accompanying schema document.
More Fields Only Help If You Can Get Answers Out of Them
A richer schema raises the ceiling on what a chief can learn from incident data. It does not change how hard it is to ask. Most departments still answer a question about their own incidents the way they did under NFIRS: run an export, open a spreadsheet, pull a second export from scheduling or payroll, and assemble the view by hand. More detail adds dimensions to that manual work rather than removing it.
That is the part FlorianAI addresses today. FlorianAI, an AI operations assistant built for fire departments, connects to the systems that already hold incident, staffing, and training records, so a chief asks a question in plain language instead of building a report. Response time distribution for one incident type over a date range. Call volume by responding unit. Overtime hours set against call volume for the same period. Every answer links back to the source records behind it, so the chief can check the number before carrying it into a council meeting.
Springdale Fire Department in Arkansas, a FlorianAI customer, is where that distinction shows up fastest. Chief Blake Holte leads 170 firefighters serving a community of over 90,000, roughly 65 percent with under five years on the job, and the department runs a fully integrated EMS service. A department running integrated EMS performs non-patient EMS support constantly, precisely the category Version 1.5 made recordable for the first time. Rising call volume, a roster where most personnel are new, and a schema that keeps adding detail all point the same way: the reporting load rises, and so does the value of being able to ask the result a direct question.
Chiefs can also ask from the assistant they already have open. FlorianAI MCP connects a department's data to Claude and ChatGPT, so the question gets asked where the chief is already working.
The Reporting Burden on Company Officers
Fire department incident reporting is consistently cited as one of the most significant sources of administrative burden for company officers. The problem is not that the reports are difficult. Most experienced officers know the content well. The problem is that completing them correctly takes significant time, the time comes at the end of demanding shifts, and the data quality degrades when officers are fatigued.
The average NFIRS report for a structure fire took 20 to 40 minutes to complete accurately. NERIS reports, with their expanded data fields and narrative requirements, run longer, especially while officers are still learning the new system. For a busy department responding to 10,000 or more incidents per year, that is a meaningful operational cost in company officer time, time that is not available for training, equipment maintenance, or the supervision and mentorship that distinguish good firehouses from great ones.
AI cannot eliminate the requirement to document incidents. But it can dramatically reduce the time that documentation takes by handling the parts of the report that are mechanical: pulling structured data from CAD, pre-populating known fields, and drafting narrative language from the incident data that the officer then reviews and approves.
How AI Automates NERIS Data Collection
A complete NERIS incident report draws from multiple data sources: CAD records, apparatus response logs, personnel assignment records, equipment use logs, and the officer’s narrative account. In a manual workflow, assembling all of this takes time even before the officer begins writing.
AI-assisted NERIS reporting works by connecting to the data sources that already contain the required information and pulling it into the report structure automatically. CAD provides dispatch time, response time, address, and initial incident type. Apparatus systems provide unit assignment and on-scene times. Personnel records provide crew assignments. The structured data fields, which constitute the majority of a NERIS report, can be pre-populated before the officer opens the report.
What remains is the narrative: the description of conditions on arrival, actions taken, and outcomes. This is where AI language capability is most useful. Given the structured data that has been pre-populated (incident type, resources assigned, timeline, outcome codes), an AI assistant can draft a factually accurate, properly structured narrative that the officer reviews for accuracy and approves. The officer’s role shifts from writing to editing, which is significantly faster and less cognitively demanding.
Data Quality as a Strategic Asset
Beyond the efficiency argument, there is a data quality argument for AI-assisted NERIS reporting. Manually completed reports under time pressure produce inconsistent data. Field descriptions vary by officer. Outcome codes are applied inconsistently. Response time calculations contain errors. These problems compound over thousands of incidents and produce a dataset that is difficult to use for operational analysis.
AI-assisted reporting, because it pulls structured data from authoritative sources rather than relying on officer recall, produces more consistent and more accurate data. That consistency matters not just for USFA compliance but for the department’s own operational intelligence. A clean, consistent incident dataset is the foundation for the kind of operational analysis (response time trends, resource utilization, incident type distribution) that supports budget justification, staffing decisions, and resource planning.
Clean, consistent incident data is only an asset if the department can question it later. That is the principle behind how FlorianAI approaches NERIS integration: the data a department generates should come back to the chief in a form they can actually use.
NERIS Integration with FlorianAI
FlorianAI supports NERIS data integration as part of its broader RMS and CAD connectivity. The specific workflow depends on what RMS and CAD systems a department uses, but the principle is consistent: FlorianAI connects to the systems that hold the incident data, pulls structured data into the NERIS report structure, and provides AI-assisted narrative drafting for the portions that require natural language.
For departments still catching up to NERIS, FlorianAI can support the move by helping officers understand the new data field requirements and mapping their existing data workflows to the NERIS structure. This is a common pain point, not that the new requirements are unreasonable, but that the mapping between old workflows and new requirements is not always obvious.
Compliance, Accuracy, and the Officer’s Responsibility
It is worth being explicit about what AI-assisted reporting does and does not do in the compliance context. The company officer remains responsible for the accuracy and completeness of the incident report. AI automates data collection and drafts narrative language, but the officer reviews and approves the final report. This is not a loophole in compliance. It is the correct division of labor between a tool that handles mechanical data work and a professional who takes responsibility for the accuracy of official records.
Departments implementing AI-assisted NERIS reporting should establish clear review protocols: what the officer is expected to verify, how corrections are made, and what the documentation trail for AI-assisted reports looks like. These protocols are straightforward to establish and are consistent with how other professional fields, including medicine, law, and finance, handle AI-assisted documentation.
What to Do Now That NERIS Is the Standard
Departments that build efficient NERIS reporting workflows now will spend less of their officers' time on documentation and produce cleaner data than departments still treating NERIS as a problem to deal with later. NERIS is already the federal standard. The risk is no longer missing a deadline; it is absorbing the heavier reporting load of a more comprehensive system with manual workflows that were already stretched under NFIRS.
AI-assisted reporting is not a shortcut around the work of moving to NERIS. It is a way to absorb the increased reporting burden of a more comprehensive system without proportionally increasing the administrative load on company officers.
For more on how FlorianAI connects to fire department data systems, see the FlorianAI overview. For a broader view of how AI supports fire department knowledge and documentation workflows, see Fire Department Knowledge Management: Getting SOPs Out of Binders.
Schedule a demo to see how FlorianAI integrates with your RMS and CAD for NERIS reporting automation.