32 pages • 1 hour read
Cathy O'NeilWeapons of Math Destruction
Nonfiction | Book | Adult | Published in 2016A modern alternative to SparkNotes and CliffsNotes, SuperSummary offers high-quality Study Guides with detailed chapter summaries and analysis of major themes, characters, and more.
Summary
Chapter Summaries & Analyses
Key Figures
Themes
Index of Terms
Important Quotes
Essay Topics
Tools
BetaChapters 4-6Chapter Summaries & Analyses
Chapter 4 Summary
Another dangerous WMD is online advertising based on data collection and assumptions about individuals. While proponents argue that targeted ads can be welcomed by consumers, these targets can be and are used for malicious purposes. One of the biggest culprits, O’Neil argues, is for-profit schools, US regularly prey upon people who are depressed, experiencing financial hardship, in an oppressed group, and with low self-esteem. They sell the promise of climbing the social ladder by unlocking the world with an education. While the idea seems noble, these schools, like Corinthian College, purposefully target the downtrodden and charge them exorbitant fees that many students will never be able to repay. Advertisers prey upon vulnerabilities, and while this isn’t unique to the online era, it’s more widespread and damaging.
These data companies learn about their targets and use their vulnerabilities against them. O’Neil argues that the most susceptible demographics are those who are ignorant about these tactics, but the other key component is somebody with a “pain point” that companies can pretend to heal or address (73). By locating exactly what ails a consumer and by knowing what they fear, online advertisers can use this arsenal of information to influence consumers into making decisions and purchases that may ultimately be counter to their best interests.
Chapter 5 Summary
O’Neil analyzes the damage caused by WMDs within the police force. She uses the example of a stop-and-frisk law implemented by the Bloomberg Administration in New York City, US disproportionately targeted Black and Hispanic young men, leading to higher rates of incarceration for these demographics due to resisting arrest and petty crimes that usually go unnoticed in other parts of society. The Supreme Court ruled that Bloomberg’s policy violated the Constitution, with the New York Civil Liberties Union calling it “racist” (93).
O’Neil acknowledges this problem isn’t isolated to one particular city. The justice system has moved to criminalize poverty by policing poorer neighborhoods and therefore catching people committing petty crimes that would go unnoticed in other neighborhoods, like minor drug offenses and jaywalking. Predpol is a well-known software used to predict where crimes will occur most, and while this software can be useful to an extent, it also demonstrates a toxic feedback loop by catching more crimes in the areas it most closely focuses on.
Chapter 6 Summary
Kronos, a company that first revolutionized payroll for companies across the country, developed a personality inventory that many large corporations use in the hiring process. At the time O’Neil wrote the book, Kronos was being sued for violating the Americans with Disabilities Act of 1990 by factoring in mental health status based on candidates’ answers into the hiring decision. Companies like Kronos developed these increasingly popular personality tests with the idea that using scientific metrics would provide greater fairness during the employment process. Often, the questions on these tests can be ambiguous with two options that are less than desirable, forcing candidates to reveal their flaws.
Proponents of personality tests might argue that a single undesirable answer isn’t enough to disqualify an applicant, but ultimately, the process isn’t transparent to candidates and often locks people who could most benefit from a job into cycles of depression. Big companies can’t be concerned about individuals that may be treated unfairly by these processes or potentially strong employees who slip through the cracks because, like many other WMDs, these tests are used on a large-scale basis.
The other dangerous element of a personality test WMD is the fact that its success is never evaluated but rather assumed. O’Neil considers some alternatives to the traditional hiring methods that often exclude minorities and women and to personality tests that may punish individuals with mental illnesses. The ideal situation would be total blindness that only focuses on an applicant’s skill, like the auditions orchestras hold behind a curtain. However, any attempts to encode fairness into machinery is inevitably coded with human bias. Instead, O’Neil argues that companies that rely on technology to make decisions should be evaluating the outcomes and improving upon them, rather than perpetuating detrimental feedback loops.
Chapters 4-6 Analysis
WMDs pinpoint the weaknesses of individuals and exploit those weaknesses for profit and control. To an extent, profit and control are linked, which O’Neil explores in this section through three domains: online advertising, crime predictions, and personality tests for job candidates. While these areas may seem unrelated, they are linked by this notion of profit and control correlating with and feeding off one another.
The online advertising WMD seeks to control a customer’s emotions in order to make them feel obligated to purchase something. By tapping into a consumer’s greatest desires or deepest insecurities, companies can unearth lots of marketing potential to help address the customer’s “pain points.” It’s not a matter of actually solving a consumer’s problem but giving the illusion that something they want most is attainable if they just had this one thing. In the case of for-profit colleges, the line of thinking goes that if the consumer just had an education, they could break out of the cycle of poverty or stop feeling stuck, and they could finally have a secure future and happiness. This manipulation—and failure to deliver—keeps these impoverished consumers buried in debt.
No company can actually sell happiness, security, prosperity, love, or any of those big nouns that fuel a consumer’s deepest desires, but they can prey upon the consumer’s fears to make them feel like they first need to purchase or invest in something in order to achieve what they really want. This is precisely how these malicious advertisers work: They feast on the insecurities of the most marginalized individuals because “[v]ulnerability is worth gold” (72). It’s not a new concept, but now that one program can target millions of people, it’s a far more damaging one than, say, the quack or snake-oil salesmen of the olden days. The profits, not the consumer’s actual outcome, is key to these advertisers because money is the end game, and as long as there are desperate, lonely, self-loathing, and downtrodden people in the world, they have a market.
This market exists because of other WMDs that work to coerce people into poverty traps, as with the crime prediction WMD. When any group is scrutinized at a higher rate, more data will be gathered about that group. For example, in the case of crime, the more police officers patrolling the streets, the more crime they’ll find because there are more resources dedicated to it. This doesn’t mean that police officers cause higher rates of crime but rather that their presence enables them to record it. Conversely, their absence from certain neighborhoods allows them to miss crimes that occur. Since police are sent to poorer neighborhoods at a higher rate, they are more likely to uncover petty crimes there. This increased rate of crime reinforces the need for the police to be there, causing a toxic feedback loop. However, similar rates of petty crime, like underage drinking, minor theft, and even jaywalking, may often go unnoticed in more middle-class and affluent neighborhoods where police don’t patrol as heavily. As a result, it appears as though there is less crime in these areas, reinforcing that feedback loop once more. O’Neil catalogs different models for handling potential crimes but ultimately comes to the conclusion that these models are self-reinforcing and perpetuate the cycles of poverty that lead to higher incidences of crime.
Another method of controlling people to yield profits is to ensure an understanding of that person’s personality before even giving them the chance to affect the bottom line. Naturally, before hiring individuals, companies like to know what kind of people they’re hiring. This, on its face, is a logical step in the vetting process that employers should take. After all, they want their employees to be a good match for their needs and to avoid any issues that may arise. It seems natural then that companies would use technology, like personality tests, to help expedite decisions about the fit of an individual candidate.
However, the use of personality tests like the one from Kronos and used at corporations across the country raises difficult ethical questions, which O’Neil teases out, such as whether personality type is relevant, whether people are being unbiased or honest in their answers, and whether those who are being honest are punished for their honesty. Specifically, these tests might single out those with mental illnesses. O’Neil’s analysis of the mechanisms of this personality testing in the workplace suggests that it isn’t always an accurate measure or a fair one but because it’s beyond question and saves companies time and money, it will persist (as long as the lawsuits don’t win out). Notably, the lawsuit from this section never reached a verdict.
The core idea in this section is that with control comes profit: Control the narrative; control consumer’s desires and perceptions; control the streets and statistics; control the workforce. Promise control to people who have unknowingly or unwillingly forfeited it, and by giving people the illusion of control, WMDs successfully covertly reap huge profits at a human cost.
