Fire agencies grapple with AI accountability as systems expand
As artificial intelligence moves into wildfire operations, fire agencies face a critical gap: the tools are advancing faster than the frameworks to oversee them. TracPlus, which manages operational data for aerial firefighting across 44 countries, released a white paper this week examining how agencies can adopt AI without sacrificing transparency or human control.
The question is direct: when AI recommends a decision in a wildfire operation, who is responsible if something goes wrong?
The scale of the problem
TracPlus tracks more than 2,500 aircraft globally and maintains over a billion flight records. The system grows by a million operational data points daily. Fire agencies including CAL FIRE, Australia's aerial firefighting program, and Fire and Emergency New Zealand rely on it as their system of record across multiple regions.
The white paper, titled "AI Makes the Call. Who Answers for It?", describes a shift already underway: AI increasingly influences fire operations while governance and evaluation frameworks continue to lag.
What the guidance recommends
TracPlus recommends three core practices:
- Run AI alongside human decisions, not instead of them
- Maintain a clear operational record of recommendations and outcomes
- Use data to evaluate AI performance over time
"AI is becoming part of the operational toolkit, and that's a positive development," said Todd O'Hara, Chief Commercial Officer at TracPlus. "The opportunity for agencies is to introduce AI in a way that preserves visibility into recommendations and decision outcomes."
For operations professionals managing these systems, the priority is ensuring that AI recommendations remain visible and auditable. Without that visibility, agencies cannot evaluate whether the tool is working as intended or assign accountability when decisions fail.
The white paper signals that fire agencies are moving past the question of whether to use AI and toward the harder question of how to use it responsibly. Those managing AI for operations will need to establish clear protocols for oversight before systems become too embedded to change course.
Your membership also unlocks:
