Artificial Intelligence and the PSAP, Can We Trust It?

Share on facebook
Share on twitter
Share on linkedin
Share on email

Sign up for our Newsletters

The National Emergency Number Association (NENA), estimates there are about 240 million 911 calls made in the US each year. These emergency calls are answered by exceptional people who can calmly, rationally, and with compassion, aid people in their time of need. AI offers solutions that could make those invaluable human response times faster as well as more informed, and result in more people receiving help.
Winbourne Consulting AI Artificial Intelligence.jpg

Artificial Intelligence or AI, is a common topic on news programs these days.  The use of AI and how it influences our daily lives includes: web searches on Google; email filters to clear out junk mail; using social media for finding long, lost friends and family members; and product recommendations from Amazon.  There are many other examples of everyday activities driven by AI where AI is being used to develop new medical treatments, driving automobiles, as well as a host of other very complex tasks.

There are many examples of how AI is used on our daily lives, but is AI being used in PSAP technology and operations?  The question of AI’s use in PSAPs will inevitably cause a discussion revolving around whether AI can be trusted with our 911 calls and the safety of first responders.
 
This article does not pretend to answer that question, but is intended to provide some insights into how AI is already being used in emergency communications operations.  This article will highlight some of the applications where AI is being used in PSAPs now.  From our experience, we know that the use of AI in the PSAP will continue to evolve, mostly behind the scene; but in some areas AI is being discussed as a tool to  supplement daily operations.
 
Examples of AI used in or by PSAPs include:
  • Natural Language Processing (NLP) is a type of AI. It is the ability of a machine to understand, analyze, and generate human speech.  NLP can recognize through voice sentiment analysis whether the caller is in real danger or pain.  For example, queue management could possibly be enhanced to prioritize calls within the queue, to ensure that “real” 911 calls are ranked higher than a “butt call”. NLP applications are already widely used.  When we ask our phones to find directions, or ask Siri the weather forecast, we are using NLP.  New 911 call taking protocol products are available using NLP to determine voice sentiment.
  • Amazon-owned home security company Ring is working with over 200 police departments that have granted Amazon access to real-time Computer Aided Dispatch data.  In this scenario, Amazon is using the CAD data to automate and improve decisions made by emergency dispatch personnel and reduce police response times.  Amazon uses a proprietary AI application to analyze the CAD data and provide input to dispatchers.
  • New call taking and dispatching products using AI that are entering the market include pre-arrival medical instructions modules.  For example, both the US-based Association of Public Safety Communication Officer (APCO) Intellicomm and Corti’sOrb product, from a Danish firm, developed call taking and dispatching protocol products using AI.
  • Both products analyze the caller’s voice audio signal, including acoustic signal, symptom descriptions, tone and sentiment of the caller, as well as background noises and voice biomarkers. These products identify distinctive features of the call automatically and search for patterns that might be useful for the dispatcher.
These and other AI-based call taking products claim to be able to process caller audio input 70 times faster than current systems.  The faster processing times will allow the software to tell the call-taker if these are repeat calls, prank, etc.  Also, voice and pattern recognition can be employed to assist first responders.  The faster processing can also assist the call-taker to recognize a potential medical problem.  An example is with some systems, the software can listen in on 911 calls, analyze the words and identify other clues that point to possible serious medical conditions.
 
Winbourne_Consulting_Emergency_911The National Emergency Number Association (NENA), estimates there are about 240 million 911 calls made in the US each year.  These emergency calls are answered by exceptional people who can calmly, rationally, and with compassion, aid people in their time of need.  AI offers solutions that could make those invaluable human response times faster as well as more informed, and result in more people receiving help.
 
We can all agree that 911 call centers have changed significantly since their introduction 50 years ago.  AI is a solution that can not only assist in routing calls more efficiently, but can even shorten the response time to help save lives.  With PSAPs continuing to suffer staffing shortages and dealing with the challenges of implementing the continued rollout of new, Internet-based technologies, AI is and can be one of the go-to tools that can provide solutions to these issues.
 
Let’s Talk!  Winbourne Consulting has 20 years of worldwide experience assisting public safety organizations in the evaluation of new technologies and products and working with these agencies as they maneuver their way through the continuously changing web of technologies.  For additional information, you can contact Winbourne Consulting at info@w-llc.com.
Scroll to Top