There are now computers in everything, or put another way, everything is now a computer. That’s how cybersecurity expert Bruce Schneier described the internet of things (IoT) at a House hearing last year following a major cyberattack on DNS provider Dyn.
As Schneier further explained in the same hearing, a phone is a computer that makes phone calls; a refrigerator is a computer that keeps things cold; an ATM machine is a computer with money inside; a car is not a mechanical device with computers, but a computer with four wheels and an engine.
And all of those computers are connected to the internet, thus the name internet of things. In a nutshell, the internet of things is the world wide web reaching beyond the virtual world of desktop and laptop computers and becoming integrated into the physical world.
The concept isn’t new; it has been around since the late 80s. However, in the past years, it has changed from theory to practice thanks to two major developments:
- Moore’s law: Processors are becoming faster, smaller and cheaper. As F-Secure’s Mikko Hypponen said at a recent TNW Conference speech, “Eventually, the chip that you can embed into any appliance which turns it into an IoT appliance is going to cost nothing.”
- Internet connectivity: As we move forward, connecting devices to the internet will become easier and cheaper as well. “Very quickly we will have internet everywhere on the planet and any device and can go online without the need for accounts or subscription or anything like that,” Hypponen said in the same speech.
The possibilities of a ubiquitously connected world are immense. As I see it, you can break down the utility of IoT into the following four categories, starting from the most trivial.
RELATED: What Mikko Hypponen teaches us about IoT security
Monitoring
One of the most basic uses of IoT is to collect data from different objects and deliver it in real time or store it for later use, such as analytical purposes.
This can be something trivial such as your refrigerator showing you live image of its contents, or to warn you when you’re running out of milk. It can be a tracker that lets parents know the whereabouts of the little ones. It can be a connected pacemaker, which allows a doctor to monitor the health of a patient remotely. Or it can be an industrial use case, such as sensors installed on the engine of an airplane, reporting various status changes to operators and engineers.
The bigger opportunity of IoT data, however is when you can combine it with machine learning algorithms. Machine learning is software that can analyze huge sets of data and find insights and actionable patterns. It has already found its way into numerous online applications and services, providing personalized experience for users (better recommendations and assistance) and business opportunities for businesses (predictive analytics, sales forecasting, automation…).
When the same pattern is applied to the physical world, unprecedented opportunities emerge. This includes predictive and preventive maintenance, a data-based approach to tending to equipment and critical infrastructure, an endeavor that can prevent disasters such as bridge collapses.
RELATED: Understanding machined learning, the general purpose technology of the future
Remote control
A more advanced use of IoT is manipulating or controlling devices remotely. In the consumer space, this can be a smart home lock that can be unlocked through a mobile application, or the combination of an internet-connected camera and a smart lock that enables you to see who’s at your door and open it for them even if you’re not present. You can control the lighting and air conditioning of your home through internet connected appliances with smartphone apps. Parents can use baby monitors to look over their children, communicate and interact with them.
In the industrial IoT, operators and engineers will be use cloud applications and web dashboards to inspect, diagnose and control devices such as traffic lights. Doctors can use connected drug infusion pumps to monitor and control the dosage of medicine for a patient.
More advanced platforms will enhance controls by providing scheduling or triggers, such as a heater that automatically turns off when the room’s motion detector reports that there’s no one left in the room.
Task automation
Putting the power of artificial intelligence and machine learning to use, IoT devices and ecosystems can shift gear toward full automation. For instance, sensors in your hotel room can monitor your usage patterns and learn your habits, such as the time you wake up, leave and return to your room, your lighting and conditioning preferences, etc. And then they can adapt themselves by automatically adjusting all those parameters without your intervention. And the data can be stored in your profile at the hotel, so the next time you return, it will already know who you are and how you prefer to have your room.
In industrial IoT, automation and machine learning can help save time, energy and money. For instance, an automated power control system can see to the efficient use of electricity. Google’s DeepMind is exploring this concept with its famed AlphaGo algorithm.
Extrapolating to other fields, there are a lot of ways that IoT and AI can make changes. For instance, in agriculture, IoT sensors and AI algorithms can help prevent food shortage crises by enabling precise farming, a practice that can help maximize yields without increasing resource usage.
The future of IoT: a fully automated economy
The future of IoT can be envisioned as smart devices that interact with each other and make decisions in such ways that no human intervention is required. Think of a washing machine that automatically orders detergent from a supplier when it runs out. The ordered detergent will then be delivered by a drone and refilled by a robobutler.
On the road, self-driving cars will be able to negotiate lanes with each other through machine-to-machine (M2M) communications and information they obtain from road sensors. Eventually, fully automated road traffic will lead to some fundamental changes, such as traffic lights disappearing altogether.
A self-sustaining economy, in which humans are less and less involved can have some radical repercussions and many things will change in nature and meaning, such as work and money. Whether this is a good thing or bad is up for debate.
RELATED: By the time self-driving cars become the norm, we won’t be needing them anymore
Challenges remain
As with every other disruptive technology, the internet of things has its own challenges. As big as the opportunities are the threats that trail behind the endless stream of IoT devices that are entering the consumer market and professional environments. Chief among them is security.
The introduction of connectivity from information technology to operation technology opens the Pandora’s box to a whole new set of security and privacy threats. The hacking of the Ukraine power grid—on two occasions—the 2016 DDoS attack on Dyn, the famous Jeep hack and numerous proof of concepts are prelude to what might come in the future. IoT ransomware, viruses that will take your car, home, office, etc. hostage until you pay up, is a threat that we will probably face soon.
RELATED: What is IoT security?
Aside from security and privacy, scale is another challenge that the industry is facing. As the number of connected increase by the billions, current networking and cloud models will be hard pressed to support the chaotic growth. Researchers are looking for various ways to deal with theses challenges, such as fog computing and the implementation of blockchain in IoT ecosystems.
There will be many more bumps on the road, but they will eventually be overcome, and the internet of things will one day become something that we take for granted, something that is an inherent part of our life, for better or worse.