AI-powered surveillance: the two sides of the coin
As with most things in life, AI-powered surveillance has two sides…
Artificial Intelligence has many uses. One of the more prominent ones revolves around the analysis of image & video material — the field of computer vision.
Perhaps the most straightforward application of computer vision is facial recognition; we’ve all seen a modern phone’s facial-unlocking capabilities. The autopilot functionality of Tesla cars is another great example of computer vision — taking decisions on the road in real-time through video & sensory analysis:
However, phone unlocking & autopilot are by far not the only applications of the field. The surveillance sector has made tremendous use of AI in recent years — sometimes for good and others not exactly.
The Good
#1. Walk-in & walk-out stores
Personally, I find waiting in line in the supermarket as one of the most tedious tasks. Wouldn’t it be nice if you could just walk in the store, grab what you want, and get out? It would indeed make life much easier and we’re not the only ones who’ve recognized this.
Startup Standard Cognition is already making software that can help with making this possible across the board!
The best of all? They don’t identify your face or sell your data!
Amazon also has a line of supermarket chains called Amazon Go — currently running in 29 locations. The idea is the same, although I wouldn’t exactly trust Amazon with my private data.
#2. Automated security camera systems
The security guards of office buildings often spend hours looking at camera screens. Even while spending so much time in this activity, a human security guard will virtually always miss an intruder.
AI systems are coming to the rescue.
AI-powered surveillance can spot patterns in behavior better than a human. Patriot One Technologies — a Canadian company specializing in automated surveillance — is leading the market in this field. Essentially, their systems can detect suspicious behavior automatically from a camera footage:
The scenarios in which this technology can be utilized are multiple: from preventing public shootings to inhibiting theft.
Automated building security systems can be a great addition to modern offices. Not only are such programs more reliable than a human in some cases, but also cheaper. However, such technological additions need to follow appropriate privacy & ethical measures, or else could quickly turn into a means of malevolent control, as we’ll see below.
The Bad
Before beginning this section of the post, I must say that I do not have a personal bias against China; it simply is the most surveillance-advanced state in the world, hence why the following examples are mainly drawn from there.
#1. Employee Tracking
Recently, PwC introduced a new AI system for its higher-level trading employees: camera tracking. The purpose of the system is to make sure that employees are at work when they are supposed to — requiring a written explanation for any absence, including toilet breaks.
While at first this might seem beneficial to the employer — more time at the workplace = more productivity — such systems actually decrease both trust & results.
Fortunately, PwC quickly received backlash on its AI-powered surveillance attempts, but the danger of this technology entering the workforce remains in place — evading personal privacy & individuality.
Not surprisingly, this does not apply to more totalitarian states; China still uses this employee monitoring technology widely.
#2. The Xinjiang province — a dystopian dream
The Xinjiang province in China — colonized by Mao in 1949 — is believed to be of key importance for developing trade routes with Europe & Asia: the so-called Belt and Road Initiative utilizes the region.
The province’s native population — the so-called Uyghurs — tend to make trouble for the Chinese government from time-to-time, mainly due to not being content with colonial existence (quite understandably):
The solution?
AI-powered surveillance cameras at every corner track the population every living second. This is a reality:
The Xinjiang province is the perfect testing playground for AI-powered surveillance and the Chinese government is certainly making use of it.
Suspicious behavior is automatically detected and instantly reported to the state authorities, who swiftly prompt an investigation; you could get interrogated for something as small as leaving your house from the back rather than the front door!
The local population lives in constant fear of being watched & stopped every time they go out. Thinking of where this could end if deployed widely is sickening to the bone: making complete mockery of free will & individuality.
Can you imagine living in a society in which you’re tracked every second? I’d rather not, but it is a reality in some places! Those are concerns which must be taken with dread seriousness…
Conclusions
As with most things in life, AI-powered surveillance clearly has two sides.
Unfortunately for those living in China and more authoritarian states, AI will increasingly become a means of control, rather than a positive addition to daily life.
However, with privacy measures in place, AI-powered surveillance can be a force of good rather than evil. It is up to you and me to keep a vigilant eye out and not allow AI to step over essential human boundaries in our own countries!
What do you think — are AI surveillance technologies a positive or a negative addition to society?
Let me know in the comments below.
StAI smart!
