Workplace Surveillance

In Weapons of Math Destruction, Cathy O'Neil discusses the philosophy behind and pitfalls of algorithms. The European Parliament defines an algorithm as a procedure used to solve problems (European Parliament 2019). In her book, O'Neil analyzes algorithms through three lenses: opacity, damages, and scale, and discusses "algorithmic decision making." Algorithmic decision making is explained as a way of electronically analyzing big data and having the machine or program use this analysis to make decisions (O'Neil 2016). Understanding of the use of algorithmic decision making is important in order to determine if there is bias within the algorithm, whether the algorithm is produced to cause harm, and ultimately to make sure that the algorithm is the best it can be.

Algorithms of Workplace Surveillance

Algorithms definitely have their influence within workplace surveillance. This is dubbed "algorithmic management." According to Data and Society, algorithmic management is "a diverse set of technological tools and techniques that structure the conditions of work and remotely manage workforces" (Mateescu et al. 2019). Algorithmic management systems have primary similarities, including:

1. Digital surveillance and data collection of workers

2. "Real-time responsiveness to data that informs management decisions

3. AI or machine learning decision making

4. Performance evaluations shifted to metric or ratings systems

5. Penalties to incentivize workers' behaviors

Algorithmic management can be see in real life scenarios. According to Lohr, within the restaurant industry, "every waiter, every ticket, every dish and drink" are catalogued and analyzed for "patterns that might suggest employee theft" (2014).

This website will discuss the algorithmic decision making of workplace surveillance through the framework of Cathy O’Neil’s Weapons of Math Destruction. Discussion will surround the pervasiveness of, potential problems of, and the reasonings or lack thereof for workplace surveillance.

In Weapons of Math Destruction, O'Neil discusses the concept of "damages" and their relation to weapons of math destruction. Ultimately, damages are exactly as they sound. Damages refers to algorithms that actively harm their users or target base (O'Neil 2016).

Taylorism

Doctorow refers to the rise of online employee surveillance as "robo-Taylorism" (2019). Taylorism is a century-old workplace management philosophy that at one time prevailed. It has since been critiqued and disregarded. Frederick Taylor's main philosophy was "scientific management," a strategy where a stopwatch was used to monitor time and a worker's every movement was catalogued. Originally seen as a way to protect workers, scientific management soon became a way for bosses to oppress. Workers came to hate scientific management. Experts like Peter Cappelli, director of the Center for Human Resources at the Wharton School of the University of Pennsylvania, suggest that this is how electronic surveillance is being used now (Lohr 2014).

There are a multitude of potential problems that come with workplace surveillance. These include lack of privacy, decreases in productivity, being a band-aid solution to real problems within the workforce, and having unintended effects and consequences.

Lack of Privacy

There are rising privacy concerns of the "specter of unchecked surveillance in the workplace" (Lohr 2014). There were and are still instances where the employees' rights to privacy were violated. An employee at an undisclosed IT services company sent a private chat, an instant or direct message, to a friend at work detailing that he was worried over potentially having shared his sexual identity with his manager and that this would affect his career. Wiretap, a surveillance tech, detected this message and, without the man's consent, sent the information to a senior exec at his company (Solon 2017). While the senior exec was able to defuse the situation, this situation still calls into question whether the surveillance here was advantageous or breached the employee's privacy.

Jason Edward Harrington worked as a TSA agent at O'Hare International Airport in Chicago for six years. While working there, his higher ups would record his and his co-workers's every moves. While this was put in place for the safety of the passengers' belongings, Harrington claims that the managers instead combed through the footage to search for the slightest infractions, eventually causing him to resign (Shell 2018). In Harrington and the TSA's case, this surveillance was misused. This is a problem that denotes privacy issues but also questions the larger culture of work.

Not only are people being recorded often without knowing the true reason, but there is also concern over who truly owns this data. When surveillance is outsourced to third-party companies, the legal ownership of this footage can get murky. "Ownership issues can affect the power dynamics of work-related surveillance in ways that cut across all socio-economic classes" (Rosenblat 2014).

Decreases in Productivity

With claims of increased productivity being the driving force behind surveillance tech, there is evidence to show that these claims may be false. According to British anthropologists Michael Fischer and Sally Applin, workplace surveillance erodes the worker's sense of agency. This in turn increases workplace stress, promotes worker alienation, and lowers job satisfaction (Shell 2017).

A Band-Aid

In his essay "In Praise of Electronically Monitoring Employees," Andrew McAfee, an MIT researcher, details a workplace surveillance experiment he conducted alongside colleagues. McAfee and his colleagues monitored waitstaff at 392 casual restaurants using theft-detection software. The installation of the software correlated with decreased employee theft and increased tips and revenue. McAfee labeled this a "win-win." What McAfee failed to take into account, though, was that it may have not been the surveillance itself that caused these benefits. It's possible that instead of the surveillance causing this increase in productivity, it was the increased attention on worker's and the environment. Anteby, a researcher, notes that "It’s possible that almost any change—even changing the lighting—would have prompted a similar increase in productivity.” It’s also possible that observed employees felt pressured to push customers to order more—a practice that is not necessarily good for business in the long run, as few of us enjoy feeling pressured to overconsume" (Shell 2018).

Instead of addressing the actual problems causing a lack of productivity, McAfee and his team focused on monitoring the employees of these establishments. While this did ultimately have a positive effect on productivity during the time of this survey, one does not know the effects on the workers' mental health or their productivity in the years to come.

Unintended Effects

Workplace surveillance can often have unintended consequences. Whether these are specifically damages is subjective, but they can definitely be potential for problems for the people monitoring.

In Britain during the 1970s and 80s, during the time of the Yorkshire Ripper murders of sex workers, British police installed security cameras in high traffic areas of prostitutes in hope to deter them from entering these areas. Instead, the sex workers used this surveillance to their advantage by staying in the sight of the cameras to make sure the cars in which they entered were recorded (Rosenblat et al. 2014).

Surveillance of this measure creates "rigid technological control over standardized work activities." This control can cause worker's to resist the surveillance in ways they would not have if they weren't being monitored (Rosenblat et al. 2014).