Blackberry has been having virtual events to make up for the COVID-19 cancellation of Blackberry World. At their latest event, they showcased how the AI technology created for autonomous cars could be used to massively transform hospitals into a far less costly, far easier to manage institutions.
Clearly, this could go beyond healthcare into airport management, retail stores, manufacturing, and other complex sites that currently require lots of people doing relatively repetitive tasks. As we face this unprecedented pandemic, maybe it is time AI took a broader role in keeping us all safe.
Let’s explore the broader application of Automotive AI this week.
What seemed like a relatively simple task of allowing a car to drive itself resulted in a massive effort to create what eventually could become a general use robot.
Initially, the auto industry thought that you could simply bury a cable in a road and have cars drive on what then would be virtual tracks along the lines of the car rides at places like Disneyland. Eventually, this evolved into creating small supercomputers that could take massive numbers of simultaneous sensor and camera inputs, resulting in the potential for computers that could out-drive humans.
Security problems were identified and overcome, a massive amount of processing power using a limited amount of energy was created, and simulation platforms were developed – all to create what should be a mass of autonomous cars coming to market over the next few years.
But all this processing power and sensor integration could be applied to other types of vehicles like trains and airplanes, personal robotics, and even buildings and smart cities. It becomes merely a matter of scale, and, at least with the fixed implementations, you don’t have to worry about automotive power limitations.
Now Blackberry has an impressive technology set applied to this opportunity. They provide binary scanning for automotive applications to assure their function and that they are secure, they design in Over The Air (OTA) update capability from a secure source, the solution included vehicle health monitoring for proactive maintenance, Blackberry protect for anti-virus, and Blackberry Persona so the car can tell who is driving it regardless of the key or fob.
Applied To Healthcare
Now imagine the systems in a hospital being able to tell who is operating them and flagging out if the operator wasn’t recognized. Or alerting if someone unknown was gaining access to patient records, assuring the software being installed did what it was supposed to and wasn’t compromised. This could provide for OTA software updates from a secured and approved sources, and could identify if a system like a Ventilator was showing signs of failure with plenty of time to correct the problem or replace it before the failure put the patient in mortal danger.
The hospital AI could identify and prioritize monitored events, identify people in the hospital that weren’t supposed to be there or who were inadequately protected, alert out if people entering protected areas were symptomatic, and could provide a hospital-wide control panel so that administrators would both know of a problem and timely. This system would even be capable of autonomous action if, during a crisis, staff were unavailable or unable to get to the patient in time.
Because the system was designed to be autonomous, it could embrace the coming wave of medical robots that would be better able to operate under these horrific pandemic conditions and the knowledge used to train one hospital could be applied through inference to other similar hospitals around the world.
And, at some future point, during critical times when staffing was short, and concerns surrounding infecting hospital staff made mitigation excessively dangerous, you could stand up largely automated emergency hospitals manned by robotic staff creating a far safer, from that staff’s perspective, environment for a highly contagious virus or disease.
Wrapping Up: Broader Implementations
Airports, manufacturing sites, food services, and even smart offices would be even easier to implement, and experimental programs like the Amazon Go automated markets (which are potentially far safer for staff than traditional markets) could all benefit from this technology. And, looking ahead 5 to 10 years, given the likelihood of both new COVID-19 infections and new pandemics, the need to automate these sites will undoubtedly drive their creation.
But I expect, as we advance Automotive AI, it will both scale up to buildings and cities and scale down to homes. NVIDIA has been telling a remarkably similar story and has already started moving into robotics, farming, and other adjacent markets.
Perhaps a general-purpose AI is a lot closer than we thought. Something to noodle on this week.