Imagine a world where live microphones and cameras are hidden in objects all around you, with powerful entities listening in, waiting for you to utter a specific word that triggers a transmission of everything you say to faraway, hidden vaults.
Is this a scene from a sci-fi fantasy set in some distant future? No, it’s the reality of consumer electronics in 2017.
Everything in the scenario described above is happening now. Everyday objects like fridges are being built with onboard computers, internal cameras, and wireless connectivity, allowing them to be linked-up to a dizzying array of microphone-enabled devices flooding the market offering voice-activated controls.
The powerful entities that are listening-in through these devices use pieces of artificial intelligence software, which are always listening, ready to awaken to begin recording when you say a prompt word out loud. The faraway vaults are proprietary servers, hidden from public view, that the AI software sends your voice commands to over the internet for processing and indefinite storage.
These devices are entering our lives at a breathtaking pace, thanks to technological progress and the insatiable appetites
of giant technology firms, consumer electronics companies, and home appliance manufacturers all jostling for emerging market share.
There may be as many as 17 billion of these devices in use today, skyrocketing to perhaps 46 billion by 2021.
Frame that differently: they’re set to outnumber humans six to one. They are the Internet of Things (IoT).
Dozens of computerized, internet- connected objects with hidden microphones, cameras, sensors and other gadgetry are already on offer today. In addition to refrigerators, smartphones and laptops, they include televisions, laundry machines, stoves, trashcans, lamps, light bulbs, thermostats, child-trackers, streaming media players, door locks, indoor and outdoor security cameras, and more.
The IoT is a double-edged sword. It has the potential to offer us unheard-of amounts of personal freedom and convenience, especially to those who would benefit from more accessible technology, such as seniors, those suffering from injuries, or people with mobility issues.
But it may also point to a dark future, where hackers, government agents, and even marketing gurus can spy on us in unprecedented ways, gleaning an alarming amount of detail about our lives, habits, preferences, relationships, and opinions through hidden sensors.
The IoT is the result of many different factors, all converging at once.
Start with computing and networking. We have arrived at the point where an object no bigger than a 5mm-thick credit card contains all the guts of a modern computer: operating system, processing power, memory, storage, and wireless internet. That means a wide-range of things can now become fully computerized and wirelessly connected.
This computing power and access to cloud-based AI can also help drive complex software that is able to analyze the output from a suite of sensors, which have also evolved. Microelectromechanical systems, or MEMS (because who wants to type that word out again?) are microscopic sensors and other elements that are practically invisible but can be crammed into different devices and hooked-up to onboard computers.
These sensors can scan a wide range of different aspects of the physical world: temperature, motion, velocity, acceleration, tilt, proximity, light, magnetism, force, torque, pressure, flow, vibration, humidity, and more.
These sensors are combined with actuators, a word meaning mechanisms that act on the environment. This technology is used for tools such as smart thermostats, which uses sensors to detect temperature, onboard computers to process this information, actuators to change the temperature in your home accordingly, and wireless internet to inform you about it. Throw in AI and it becomes self-learning, adapting to your behaviour as the seasons change, or as you demonstrate that you like to eat your breakfast in a climate of 22 degrees C.
All this tech together makes for devices that are truly autonomous agents: computerized, AI-driven, loaded with sensors, connected to mobile internet, and battery-powered with no need to recharge for a long time. (Throw in self-driving cars and drones, and…well, you get the picture. Our robot future is suddenly upon us.)
The benefits of these advances may not readily be apparent, especially for things like home appliances. Who cares if I can talk to my fridge? Plus, gadgets like coffeemakers, for example, have long been electronic. I’ve been able to program my coffeemaker to turn on at 6 a.m. since the 1990s.
But having all of this new technology is a whole different ballgame. It’s the difference between your coffeemaker being simply programmable, and your coffeemaker talking to your cell phone, alerting you when your water level and stock of coffee is running out, and then ordering more to be delivered with the swipe of a screen.
Think of the benefits that provides, say, an office manager taking care of their break room at work, or a small business owner running a chain of fast-food restaurants.
Similarly, IoT devices allow for a much more refined level of remote environmental analysis and control, for those wanting to stay in command of a space’s energy use, security, access, lighting, ventilation and the like, all over the internet via their phone or tablet. All sorts of folks in the business and personal worlds could benefit from this remote control of physical space and the big data that it will bring, which when coupled with machine learning, can help save on electricity bills or provide peace of mind for property owners.
There are even greater potential benefits to humanity: the IoT can be a dream come true in terms of accessibility. For some of us, using phones or computers, operating our home appliances and gadgets, and being productive in the workplace are all second-nature to us. But for those of us with disabilities, injuries, or mobility issues, these kinds of things can be difficult or impossible to accomplish.
The most obvious example of this is the rollout of Intelligent Personal Assistants, or IPAs (no, not the beer style). Google’s Assistant, Amazon’s Alexa, Apple’s Siri, and Microsoft’s Cortana are all IPAs. They are essentially powerful AI software that live on the cloud and are beamed into your life through wireless devices with embedded microphones. They respond to voice commands, then connect to the internet to carry out whatever task is being asked.
Because they respond on command, IPAs are always on. In order for them not to be constantly triggered by random conversation, they wait for you to speak magic–er, “wake words,” which are keywords that you say that make the device accept commands.
For example, you have to say “Hey, Siri” to your iPhone in order for Siri to respond, or “Alexa” to get Amazon’s assistant to perk up, or “Ok Google,” and so on.
Users who, for whatever reason, can’t or won’t operate a cell phone or computer, or who must expend a great deal of effort to do so, can now use IPAs to do all sorts of things: order items from e-commerce websites, make to-do lists, control music or video, order food for delivery, set alarms, read the news out loud, call taxis to pick them up, make reservations, give them directions, tell lights in the home or appliances to turn on or off, and so on.
The possibilities are endless, and the list grows longer every day as more IoT devices enter the market. It isn’t even necessary to use voice commands: more and more things can be done through devices with sensors that recognize gestures, like waving your hands in a certain way.
Health care and aging is another whole area where the IoT can help increase accessibility. One example is pill bottles that glow and make noise when people need to take their medicine, and that call the pharmacy for refills by pushing a button on the lid. There are also insoles that vibrate a side of your foot to let you know which direction to walk, helping the blind or the vision-impaired navigate.
But simmering under the surface of these leaps in technology is a hidden war (isn’t there always?) There’s a rivalry for market share between the technology giants Alphabet (Google’s parent company), Amazon, Apple, and Microsoft.
All of those companies have developed their own IPAs. But most of those companies have also, slowly but surely, built relationships with manufacturers that allow them to deploy these assistants into your personal life.
Google’s mobile operating system Android, for example, is in millions of phones made by companies like Samsung; while Microsoft’s Windows is in millions of PCs thanks to makers like Dell.
Now Google and Microsoft can simply bundle their assistants with the latest version of Android or Windows, which get put on Samsung and Dell products, and boom—they’ve reached the millions of consumers who buy those products.
Apple has taken another route, building its own loyal following of iPhone, iPad and Mac users, which all run Apple’s own proprietary iOS and OSX operating systems. Apple can simply put Siri into the latest versions of those systems.
Amazon, meanwhile, which has become an e-commerce and cloud computing behemoth, was still being forced to piggyback on other platforms to deploy its AI assistant, Alexa. It had partially addressed this issue by coming out with the Amazon Echo and Echo Dot, which are internet-connected pieces of hardware in which Alexa lives. But that still limits the deployment of Alexa to homes that have these devices.
That’s why Amazon is partnering with appliance giant Whirlpool, for example. If devices don’t house internet and embedded microphones to power Alexa themselves, like GE’s new lamp, they can communicate over WiFi or mobile internet to an Alexa- enabled device controlled by the same owner.
That gets Alexa to start living in more and more things in our lives. Already one news outlet has declared that Alexa is “the future of computing” as commodity giants prepare to install her into everything from refrigerators, to lighting, to the newest line of Ford vehicles.
This suggestion of domination brings us to the potential pitfalls of having all of these smart monitoring devices hooked-up to our goods, measuring us constantly, and automatically sending that data far away.
Recently, a six-year-old girl in Texas used Echo to order a $170 dollhouse and four pounds of sugar cookies. That in itself is a bit of an unsettling, if amusing notion: without the proper security controls, your savings could be plundered by accidental purchases.
But here’s where the IoT comes in. A California television channel was covering the dollhouse and cookie incident, and during the on-air segment, one of the anchors commented on it, saying “I love the little girl saying, ‘Alexa, order me a dollhouse.’”
When he said that, it was broadcasted into TV sets across the state, and when some of those TVs played the sound of him saying it, an army of Alexas perked up, because they heard their names.
When they processed the rest of his statement, they understood it to be a command to purchase a dollhouse themselves. The orders flooded in. Ordering products with voice commands, it turns out, is the default setting.
The result of that type of hijacking—mass orders of dollhouses—is financially damaging and comes with legitimate privacy concerns, but is relatively benign in terms of public safety. However, it underscores that these devices are always listening, and are recording what we tell them, for indefinite periods. Police in Arkansas, for example, have already asked Amazon for recordings made by its Echo devices for evidence in a murder case.
Besides that, though, there are other types of hijacking that are even more serious. A cyber attack this past fall on Dyn, a company that facilities internet connectivity, took out big services like Twitter, PayPal, Reddit, SoundCloud and others, including news sites like the New York Times, by taking over 380,000 IoT devices.
The attack was a fairly common Distributed Denial of Service attack, where a swarm of many machines are ordered to access a website simultaneously, overwhelming it and knocking it offline.
The analog equivalent would be getting a bunch of people to phone an office
building, over and over, tying up the lines so that when other people phone, they just get a busy signal. Traditionally these attacks have been carried out using networks of internet-connected computers, but now it has been shown that it can be done using IoT devices.
This shows how weak the security around IoT devices can be. The company that manufactured
the circuit boards used in many of the devices involved in the Dyn cyber attack, for example, blamed users for not changing default passwords. The software that the hackers deployed automatically hunted around the internet for any such devices that had easily-guessable passwords like “12345.” It had a big list of these codes, and it just went from door to door, so to speak, trying them all.
The IoT holds great promise for ease of use, connectivity, accessibility, mobility and convenience, but its deployment in every aspect of our lives over the coming decades should also be met with a skeptical eye. It means we’re losing more of our privacy and are being tracked in more ways in more places. And it means the devices closest to us, in our bedrooms, pockets, cars, kitchens, and living rooms, are connected to computers far away and are vulnerable to exploitation and manipulation.
Welcome to 2017.