How smart devices collect and use data about you every day
Embedded devices – those with computers embedded in them, such as digital watches, TVs, smartphones and, of course, personal computers – are prevalent in today’s world. They come in many forms, from simple digital controllers on microphones and fridges to more complicated home alarms, smart lighting and devices that track biometrics, such as Fitbits.
These devices – while exciting, sophisticated and convenient – pose an interesting challenge, technologically and ethically. To understand this, let’s think about what happens with the data these devices currently or may, one day, collect.
Your smartphone knows your location, which apps you use, which people you talk to, when you are on your screen, what your purchasing habits are, how long your attention span is, and so on. Your smartwatch knows your heart rate, your workout habits, your physical health. If you use a fingerprint scanner or face recognition to log into one of these devices, they also have those biometrics. Even your microwave and fridge could, in theory, track when you use them and figure out when you eat, and so be able to determine at what times of the day you might be most susceptible to advertisements about food. Think about what happens when people start tracking and analysing this data to target you in various ways: the friends Facebook recommends, the events Google shows you, the books Amazon recommends, the restaurants Siri suggests, and more. Your entire world can be subtly but powerfully altered by algorithms made to influence or make money off you.
This is scary, but fortunately it’s quite difficult and complicated to put together all the data that these tracking devices are uploading. There is a lot of it, it may vary across days, there may be noise in the data, the tracking technology may malfunction, and so on. Even if the data could be analyzed, it is always uncertain what it is saying, and so it’s quite difficult to know what to do based on the data. So even though Amazon knows about your entire purchase history on its catalogue, and Netflix has your entire viewing history, their recommendations aren’t spectacular. This is because Netflix doesn’t know what’s happening in your head when you’re watching something – you may find it funny, scary or interesting, or be bored but continue watching it because your friend wants to. Because of this, and because your tastes may change, it has a difficult time tailoring its recommendations in a way that always matches what you want to watch on a given day. The more variety in your life, the less Netflix can pin down what you may want to watch. Until someone has data analysis so sophisticated that they can figure out what’s happening in your mind, all the data collected on you will not be able to capture your motivations perfectly.
Orwellian concerns aside, there are two more serious ones worth considering. How much should these algorithms be able to “know” about people? Should an algorithm be able to know things about you that you haven’t consented to have analysed? Currently, legality and privacy settings around this are not very well formulated, and companies get away with a lot. Even if you’re okay with a company like Facebook knowing about you, there is still the secondary concern of third parties such as hackers (Facebook had 30 million accounts breached recently, through no fault of the users) obtaining this information and using it to steal your assets or sell your information to advertisers or more malicious propaganda campaigns. It’s great that the Fitbit on your wrist can track your every step, but so could a stalker who hacks into that database.