Vass Bednar is the executive director of McMaster University’s new master of public policy in digital society program.
You may have heard about cameras that constantly monitor someone’s movements at work, and clocks that document the time taken to complete every task. Though this extreme surveillance may seem like something that is only happening overseas – think Foxconn factories in China, where workers produce Apple products – human beings being supervised by AI is increasingly a distressing reality for low-wage retail and service workers here in Canada.
While many professionals are working from home and casually ordering food and packages from their laptops, these white-collar workers are so insulated from these “optimizing” practices that they may take for granted that they can be 10 minutes late for work, surf the web now and then, or go to the bathroom whenever they need to without penalty.
It is striking that we have set employment standards around when people can work and for how long, what breaks they can take, how many sick days they are entitled to and what vacation is warranted, but we have nothing in labour law that protects people from workplace technologies that automate tasks and force an unhealthy pace of work through algorithmic manipulation. This occurs when algorithms oversee the efforts of human workers to make data-driven decisions that can lower costs and achieve greater efficiencies, typically to the detriment of employees. For instance, Amazon warehouse workers are managed by algorithms that determine which items need to be moved in a specific amount of time (called the “pick rate”), and the Deliveroo platform expects drivers to accept new customer orders within an average of just 30 seconds.
A recent book from American journalist Emily Guendelsberger, On the Clock: What Low-Wage Work Did to Me and How It Drives America Insane, chronicles a working life dictated by these robotic regimes. Over the course of the memoir, she worked at an Amazon fulfilment centre, a call centre and a McDonald’s. Each employment engagement placed obscene demands on her time: how little she had for a break, how many tasks she had to do in a certain amount of time. These “goals” were facilitated by algorithms designed to push low-wage workers to the absolute brink. A form of optimization, these technologies treat people as robots, not humans, and there is little to no respite. Ms. Guendelsberger calls them “cyborg jobs.”
Many of Canada’s essential workers are being treated as cyborgs subject to the same relentless expectations imposed in digitally mediated work, whether they are working a brutal megashift at an Amazon warehouse, delivering an UberEats order or servicing a drive-through. Among an algorithm’s primary tasks is optimizing to keep staffing levels as low as possible, and so creating hectic and stressful work days. What’s worse, these cost-cutting computer programs introduce psychological burdens, physical stress and cause injuries. Their new ubiquity means workers can’t vote with their feet because the applications quantifying workers’ job performance with metrics, and the lack of unionization within and across these industries, makes it nearly impossible to avoid.
A conversation about developing more sophisticated policy interventions that can limit this unreasonable surveillance is overdue in Canada. As part of a broader policy strategy to rein in unchecked technology companies, we need new labour legislation that prevents companies from excessive algorithmic manipulation of workers that results in frantic shifts, unpredictable work schedules and a loss of autonomy.
To that end, the state of California recently introduced a bill to regulate quotas for warehouse workers to make sure they are not too onerous. The bill is intended to specifically address warehouse worker injuries. In Canada, a recent Toronto Star investigation found that Amazon’s injury rate in Canada is 15 per cent higher than the U.S. average. While “fair workweek“ laws such as the 11 Hour Rule in Ontario can help protect against unfair scheduling, technology’s reshaping of service work demands a more holistic approach.
Workers have resorted to trying to manipulate the very programs that dictate their work: recently, more than 26,000 DoorDash workers in the United States organized on Facebook to decline all low-paying offers in an effort to “game” the algorithm that determines their payment (#DeclineNow). These workers have said that the effort is increasing their average pay per order. In contrast, provincial labour standards do nothing to protect people from excessive digital task management.
Collecting as much data as possible from workers creates a new kind of inequality that gives employers more power, and employees less control. While some of these surveillance tools may be intended for more legitimate means – e.g. protecting trade secrets – most of them erode the workplace and harm workers in the name of profit.
In the meantime, en-vogue “future of work” conversations focus too much on building skill sets that insulate workers from being replaced by automation, and not enough on resisting and rejecting the trends that are already reshaping everyday work for the worse. While the economy is changing and workers are whipped through tasks, our public policy discourse is ignoring how algorithms have come to govern low-wage service work. While we often explore the prevalence of “gig” work with curiosity, we understudy how artificial intelligence, algorithms and automation are hurting workers.
Our federal government has an ambitious agenda to rein in Big Tech through a digital sales tax, new content standards and potentially forcing tech giants to pay for news. These interventions can create accountability currently lacking in our technology ecosystem. We cannot continue to ignore how companies use technology to squeeze workers and micromanage their time.
We have a massive opportunity to innovate and modernize labour standards to protect workers from dangerous algorithmic creep and make this kind of work more humane. The use and abuses of these technologies is well-documented. It’s time for provincial labour legislation to take on algorithmic management in the workplace.
Your time is valuable. Have the Top Business Headlines newsletter conveniently delivered to your inbox in the morning or evening. Sign up today.