Amazon Accused Of Using Algorithms To Push Warehouse Workers To Breaking Point

Amazon has used tricks, algorithms, and surveillance to discourage warehouse employees from unionizing, according to a paper published in the journal Socius.
The study, titled "Weaponizing the Workplace: How Algorithmic Management Shaped Amazon's Antiunion Campaign in Bessemer, Alabama," was conducted by Teke Wiggin, a researcher at Northwestern University in the US. It's worth reading the full write-up if you have time; we'll summarize some of it here.
It focuses on the 2021 union vote at an Amazon warehouse in the city of Bessemer, Alabama – population 25,000 – which ended with employees voting against unionizing. The report claims this result can be attributed in part to Amazon's use of algorithmic management practices.
Make 'em feel better ... for now
"The findings reveal that employers can weaponize elements or effects of algorithmic management against unions via repurposing devices that algorithmically control workers, engaging in 'algorithmic slack-cutting,' and exploiting patterns of social media activity encouraged by algorithmic management," the paper says.
"Algorithmic slack-cutting" is a term the author uses to describe the softening of the "electronic whip," a characterization borrowed from prior research on algorithmic work demands. It's the proverbial carrot as opposed to the stick – removing the mental burden of being under automated, software-driven oversight. The argument goes that basically, it's such a relief to a worker when it's removed that it feels like a benefit.
A cited example in the study comes from a warehouse worker's Time-Off-Task tracking system – which tracks the number of minutes that workers aren't actively working – being disabled to win over workers during the union campaign. This allowed workers to take bathroom breaks without being on the clock, ostensibly to soften attitudes toward management.
The report also describes the equipment used for what prior research characterized as Amazon's use of "coercive control mechanisms." It touches on, for example, the scanners Amazon workers use to register products to be shipped, devices that direct workers to pick, sort, or pack, and which can be used to audit and discipline them.
... to publicly rank them, and to algorithmically monitor their compliance with quotas ...
"In all, interviewees confirmed Vallas et al's description of scanners as enabling management to generate real-time data on the productivity of workers, to publicly rank them, and to algorithmically monitor their compliance with quotas, thereby imposing 'the constant threat of discipline and termination on workers who fail to make rate,'" the report says.
As an example of this, the report shows screenshots of workstation displays that present warehouse workers with questions that interviewees deemed to be coercive. The concern was that multiple-choice questions presented to workers, such as "Does your manager care about you as a person?" might be detrimental if answered negatively.
Crucially, "during the union drive, some workers believed that these questions were adjusted and/or used to gauge union sympathies or purge disliked managers en masse," the report explains.
Amazon insists: Safety checks, not snooping
A spokesperson for Amazon challenged the study's suggestion that the multinational mega-corp misuses technology to snoop on and manipulate workers:
The report also details how Amazon used its A to Z app and a separate messaging tool during the Bessemer unionization drive "to send anti-union messages to workers."
The A to Z app, available for iOS and Android as well as the web, is used by Amazon workers to clock in and out, request time off, submit HR reports, look up pay records, and more. It can ping people with notifications on their phones, which are optional, but supervisors are instructed to pressure workers to enable these, it is claimed. According to the study, Amazon used these notifications as well as the separate text messaging app to bombard people's smartphones with anti-union propaganda.
Amazon's spokesperson did not directly address the claims of pro-management lobbying, stating instead:
Amazon questions sample size
Wiggin's work is based on "42 in-depth interviews with Amazon workers and analysis of court records obtained through a Freedom of Information Act (FOIA) request."
Nonetheless, Amazon's public relations, known for its pushback against unwelcome reporting, takes issue with that claim and the report in general. In response to our request for comment, an Amazon spokesperson referred to the "study" in scare quotes.
Amazon's spokesperson went on to suggest that of those interviewed, only 34 were eligible to vote in the union election, questioned the lack of published proof that those interviewed were employees, and noted those interviewed represented only a tiny fraction of the warehouse staff.
In response, Wiggin said: "The workers were wearing their Amazon badges and uniforms or sent me pictures of them. I was in Bessemer right near the warehouse and visited the entrance a number of times. Are they actually questioning that? Forty-two interviewees is a pretty good sample for a context like that. Sixteen of my interviewees voted against the union. Much of what I report is corroborated by court transcripts and exhibits; these are cited in the paper."
Wiggin said the paper reflects four years of work, in large part because it's academically rigorous.
"Amazon implements algorithmic management in the most comprehensive way we've seen so far in a certain respect," Wiggin told The Register.
"That's because it is imposed on workers and a physical workspace. There are extra elements of direction that aren't present in gig work. Amazon workers are picking or are stowing or are packing and they're getting their instructions to do that through these workstation displays and scanners.
- Amazon workers are in a warehouse of pain, independent report finds
- On one Prime Day, Amazon warehouse workers endured '45% injury rate'
- Warehouse safety citations could cost Amazon seconds in revenue
- What price your home delivery? Amazon accused of hiding real injury rate in its overworked warehouses
"So that's all being administered through algorithms and that feels like a more intensive form of algorithmic management than, say, being an Uber driver where you're not there in person, with this physical infrastructure like directing your work and setting your pace."
'Electric whip' linked to injury rates, study claims
Wiggin argues that Amazon's electronic oversight, or as the paper terms it, "electric whip," deserves special scrutiny because it involves directing workers' physical activity.
As a result it's more dehumanizing. And this is why we see such higher injury rates
"The physical movements of Amazon workers, which are directed through these algorithmically generated quotas and enforced through automatic termination or discipline, are just destroying these people's bodies," he claimed. "So it's pretty simple, it's just applied to physical movements in a way that we don't see at other employers that use algorithmic management and as a result it's more dehumanizing. And this is why we see such higher injury rates."
Various reports have found high rates of injury among Amazon workers compared to rivals, an issue that Amazon happened to address on Thursday in an update on company efforts to improve workplace safety.
Asked what kind of labor reform would be appropriate, Wiggin questioned why there are union elections to support worker rights in the US when countries in Europe offer those rights by default. "In Europe, you see representative mechanisms that are baked into the workplace," he explained. "It's assumed that people should have control over their lives."
That's not likely to happen any time soon at the US federal level, according to Matt Scherer, senior policy counsel for workers' rights and technology at the Center for Democracy & Technology.
"It's been a long time since Congress has passed meaningful regulation advancing workers rights," Scherer told The Register.
"Arguably, the last really major labor rights law that was passed was the Occupational Safety and Health Act [in late 1970]. The trend at the federal level for a long time has been to weaken labor protections rather than to add new ones."
The trend at the federal level for a long time has been to weaken labor protections
A recent study by the Center for Democracy & Technology (CDT) examines what workers think about algorithmic management and digital surveillance, which have become pressing concerns with the emergence of the gig economy and the rise in remote work.
"Companies frequently use surveillance and automated management systems to accelerate the pace of work, often to levels that threaten the health and safety of workers," the CDT report says.
"Some gig economy platforms use algorithms that drive down wages and lead to unpredictable pay patterns for workers. Workers are increasingly held to opaque and arbitrary performance standards and face discipline or dismissal for failing to meet them. These trends have continued to accelerate even though studies indicate that electronic monitoring does not improve employee performance and may even lead to counterproductive work behaviors."
According to Scherer, the survey of 1,800 people, followed by deliberative polling involving 170 respondents to explore the issues in more detail, yielded surprising results.
"Workers aren't just like, no, we think that electronic monitoring should be banned, or, no, we think that employers should never be allowed to collect our data," Scherer said. "One thing that was interesting is that [between the initial survey and more detailed follow-up deliberations with respondents], workers became less extreme. They definitely preferred more nuanced approaches.
"My takeaway for employers, and for people who advise employers, is if you explain to workers why you're doing this, and you give them some agency and input into the process, at least you won't be undermining trust."
Scherer said that while he's hopeful employers recognize the value of extending some trust to employees, he understands that businesses ultimately prioritize making a profit. And so he said he hopes the states will impose some common sense guardrails that ensure employees are told about information collection and monitoring, that prohibit tracking outside of work or in ways that could threaten employees' health and safety. ®
Speaking of Amazon... The internet titan this month sued the US Consumer Product Safety Commission (CPSC) in federal court in Maryland [PDF], accusing the watchdog of going outside the law by making the megacorp responsible for safety recalls of stuff sold via its online marketplace by others. Amazon reckons it should be treated as a neutral logistics provider by the commission, not as a distributor with more legal liability.
This comes after the CPSC ordered Amazon last year to recall hundreds of thousands of products sold by others via its souk. The tech giant would rather not be subjected to such demands, hence this lawsuit.
From Chip War To Cloud War: The Next Frontier In Global Tech Competition
The global chip war, characterized by intense competition among nations and corporations for supremacy in semiconductor ... Read more
The High Stakes Of Tech Regulation: Security Risks And Market Dynamics
The influence of tech giants in the global economy continues to grow, raising crucial questions about how to balance sec... Read more
The Tyranny Of Instagram Interiors: Why It's Time To Break Free From Algorithm-Driven Aesthetics
Instagram has become a dominant force in shaping interior design trends, offering a seemingly endless stream of inspirat... Read more
The Data Crunch In AI: Strategies For Sustainability
Exploring solutions to the imminent exhaustion of internet data for AI training.As the artificial intelligence (AI) indu... Read more
Google Abandons Four-Year Effort To Remove Cookies From Chrome Browser
After four years of dedicated effort, Google has decided to abandon its plan to remove third-party cookies from its Chro... Read more
LinkedIn Embraces AI And Gamification To Drive User Engagement And Revenue
In an effort to tackle slowing revenue growth and enhance user engagement, LinkedIn is turning to artificial intelligenc... Read more