LO
Emergency dispatcher — tasks, one by one
The unit of analysis is the task, not the job title. Each one below carries its direction, whether the judgement rests on evidence or on platform inference, the reasoning, and what it does not establish.
Every task on this page#
Answering and triaging calls
Being augmented✓ Evidence-backedAnswering emergency and non-emergency calls, working out what is happening and deciding how urgent it is.
AI now handles part of the non-emergency calls: a US county's system resolved 2,920 of 9,635 in one month without a call-taker, and England's police 101 line uses AI to send non-police calls to other agencies before they reach a handler. Emergency lines are different: the UK says the 999 system has not changed, and in a Danish trial AI alerts did not help dispatchers recognise more cardiac arrests.
Non-emergency lines in two countries and one trial; they do not show AI answering emergency calls, or how many call-taker hours this saves.
Locating callers and recording incidents
Being augmented✓ Evidence-backedFinding out where the caller is, entering the incident into the dispatch system and understanding callers who speak another language.
Machines increasingly supply what dispatchers used to ask for: US carriers must route wireless 911 calls by the caller's location, so fewer calls arrive at the wrong centre, and a New Orleans centre shows callers' words translated on screen. People still confirm the location and, in New Orleans, still bring in an interpreter to ask further questions.
One US rule and one centre's tool; they do not measure how much questioning or data entry has changed.
Dispatching and coordinating units
Still human-led≈ Platform inferenceChoosing which police, fire or ambulance units to send, tracking them and coordinating between agencies.
No record found shows AI choosing which units to send in live operations; the decision stays with dispatchers.
The absence of a record; it does not rule out research systems that rank calls or suggest resources.
Giving instructions until help arrives
Still human-led✓ Evidence-backedTalking callers through CPR, bleeding control or getting to safety, and staying on the line with them.
Instructions stay with people: in the Danish trial, AI alerts did not change how often or how quickly dispatchers started CPR instructions, and the New Orleans centre still gives emergency instructions through a certified interpreter rather than machine translation.
One trial and one centre's practice; they do not show what callers or outcomes would be with automated instructions.