Think Artificial Intelligence or AI and the first thing that comes to mind for most is a machine that can walk, talk, think and most importantly, make decisions just like humans do. And while we may be decades away from having anything like what we see in the movies, Andy Stanford-Clark of IBM says there is no denying AI’s presence all around us.
“AI comes in a lot of guises. For example, a lot of people talk to their phones or virtual assistants. The car I bought recently has features like automatic braking, it reads speed signs and alerts me if I am going too fast. All these are examples of how machine automation is taking over the role of what intelligent humans do, and therefore can be considered as a form of intelligence.”
So how does one distinguish between smart machines which we are exposed to in almost every facet of our lives, and those which have AI elements? IBM uses the term cognitive intelligence to distinguish between the two.
“You are training it to analyse, theorize and actualize like we do. We observe our environment; we evaluate it and we have an idea about how to react to it. We then try out that idea, and if it doesn’t work out the way we expected it to, we try something else. This interactive loop of observing and learning is what distinguishes mechanised behaviour from AI.”
Providing AI with eyes and ears
But having intelligence by itself is not enough. You also need to figure out how to provide this intelligence with information to process. And that Andy says, is where the Internet of Things or IoT comes in.
“If you had a brain in isolation in a jar, it would be hard to know if it is intelligent, because it wouldn’t be interacting with anything in any way in the outside world. It’s the same with AI. You need to provide it with information, with data and one way to do that is through the IoT, which acts as the eyes and ears of the AI, allowing it to understand the world around it, make a hypothesis and then do something, like turn off a pump, turn down a thermostat, etc. which is what AI is all about, the ability to use machines to do tasks so that humans don’t need to.”
And that may be the biggest challenge that AI faces today. While there are various devices built around IoT, from refrigerators to toothbrushes to lighting systems, the killer application for mass consumer use of AI has not yet been found. The tendency so far has been to integrate AI and IoT into devices just because people can, rather than integrating it to solve a genuine problem. And while that search continues, Andy says, the future of AI could lie in its industrial applications.
“If you look at the work we are doing with KONE, on predictive maintenance for example and similar projects in other industries, these are projects with less risk. If we can predict maintenance problems, to tell when something is going to fail rather than service something once a year, or wait and then react when it breaks down, the saving in terms of cost and time can often pay for the investment in IoT and AI technologies on its own.”
Taking AI Mainstream
KONE and IBM are collaborating on the 24/7 Connected Services project, where IBM’s Watson IoT platform is used to allow these machines to talk to the cloud in real-time and intelligently. This information is then used to improve efficiency, predict breakdowns and offer more services to KONE’s customers. And according to Andy, this is just a taste of things to come.
“First, it generates a lot of efficiency and allows for better service, but as we move on, collect more data, get more infrastructure in place, we can turn up the knob on AI and do some really smart things about the way people interact with elevators, for example, sensing when someone is leaving their hotel room and having the lift standing ready. We are seeing huge quantifiable benefits from applications like these, which will then propel AI to be integrated into applications in other industries as the technology improves.”
Examples such as predictive maintenance or even the ability to automatically turn the thermostat down showcases the manner in which AI can effect genuine change, something that Andy says will make it acceptable to the mainstream and allow for more applications to be built around it.
“I have seen a lot of AI already in use in trading applications which are used to second-guess the stock markets. We are even using it in healthcare, in oncological research programs where IBM Watson is used to help guide the clinician to diagnose various types of cancers. I think finding applications that will have a genuine use in people’s lives is what will drive AI usage in the world ahead.”
And that progression is almost inevitable as the pace of technology continues to accelerate. Each year sees more memory and computational power being offered for the same cost, which in addition to cloud technologies, offers virtually unlimited capabilities which Andy says will allow future AI to take on more advanced roles.
Future applications of AI
What are some of the applications that we will see AI being put to in the near future? Andy foresees multiple uses including increased integration into people’s homes, with AI controlling many of the appliances and systems such as lighting, water use and security. Our driving experience is already being augmented with self-parking and automated route finding now a reality, and eventually we will have self-driving vehicles. Andy also expects the healthcare sector to benefit greatly, with AI helping predict potential health problems, performing predictive maintenance for our bodies. Cognitive technologies will also see industries being able to offer better services and products at a lower cost, while also delivering superior customer service.
How are we going to see AI integration play out in the coming years? Andy feels that it will be a collaborative undertaking.
“Small business who are willing to try AI technologies will be the first to drive innovative new applications. Big companies like IBM will provide the investment and market presence to move things into the mainstream, while universities will provide the R&D for future technologies. But most importantly, it will be ordinary people who will use these products, give us feedback and make sure that a product is not seen as Big Brotherish.”
With technology and computational power advancing at a rapid pace, the scene is set for a perfect storm. So get ready for AI bringing in a radical change to our lives, unlike any seen before.