The Algorithm

November 1, 2023
A disturbing investigation into use of AI systems in hiring, firing, and employee surveillance. As Emmy-winning journalist and journalism professor Schellmann demonstrates, AI has moved into crucial areas of our lives, but the process has been so fast and silent that its influence is almost invisible. The author is particularly interested in how it has infiltrated the business world, especially how it affects recruitment and dismissals. The use of AI systems began as a means to help human resources managers sort out the huge numbers of applications they received, but once the genie was out of the bottle, it spread into all areas of assessment, looking for keywords and patterns. These systems are already ubiquitous, writes the author, used by 99% of Fortune 500 companies. Schellmann records many cases of bias on the basis of gender and race; even a person's zip code and social media posts can lead algorithms to jump to specific--and sometimes incorrect--conclusions. Bias is unintentional, but it does not seem possible to give algorithms a sense of context for making decisions about applicants. The results provided by an algorithm look impressive, but follow-up research has shown that AI-based hiring is no more successful than traditional methods. Comprehensive surveillance of employees is now possible, with AI systems tracking computer use and interaction with other employees. Schellmann examines instances of effective employees being dismissed simply because their results did not match an algorithm's metrics. She argues that HR managers should be required to understand how their algorithms work, and there must be greater human input to personnel decisions. The author presents numerous good ideas, but she concludes that "it's a dark outlook--a system in which algorithms define who we are." This eye-opening book makes it hard to disagree. With clear-minded authority, Schellmann uncovers a fraught, often unfair system.

COPYRIGHT(2023) Kirkus Reviews, ALL RIGHTS RESERVED.

November 20, 2023
Schellmann, a journalism professor at New York University, debuts with a disquieting survey of the failures of artificial intelligence used in corporate personnel decisions. Pushing back against claims that AI reduces bias in hiring, she notes that AI software’s “one-size-fits-all” approach often marginalizes people with disabilities. For instance, an AI game claimed to test job candidates’ “processing speed” and creativity by measuring how quickly applicants could hit the space bar, putting people with motor disabilities at a disadvantage. AI, Schellmann explains, frequently proposes ludicrous correlations because, in analyzing the résumés of current employees, they often pick up on statistically significant but arbitrary commonalities, as when one hiring tool “predicted success for candidates named Thomas or Elsie.” Stories of people negatively affected by AI exasperate, such as the case of a recent college graduate who worked as a contract delivery driver for Amazon during the pandemic until a technical glitch triggered the automated management system to fire her. Elsewhere, Schellmann’s reports on testing various AI programs provide amusing anecdotes about the technology’s considerable shortcomings (an automated interview program designed to assess English proficiency decided Schellmann’s English was “competent,” despite her answering entirely in German). It’s a striking indictment of AI’s flaws and misuses. Agent: Roger Freet, Folio Literary.

December 1, 2023
Schellmann, an investigative reporter and journalism professor at NYU, examines the use of artificial intelligence in the workplace. There are currently many forms of AI-based software being used by employers at every phase of the employment process, from screening and hiring applicants to determining promotions and terminations to "optimizing" employee performance. Schellman's painstaking analysis includes real-world testing of some of these products, many of which make outlandish claims of accuracy that are not supported by evidence. AI job interviews, performance assessments, personality typing, and predictive tools are portrayed in this book as a grand experiment in AI that is not ready for prime time. In fact, the book makes a convincing case that many of the key players involved in the process do not themselves have a full understanding of the tools' limitations or the enormous potential for negative outcomes. The author includes tips for workers and job seekers trying to adapt to this new reality. A stunning portrait of the misuse of technology in the workplace, this book is a must-have for general business collections.

COPYRIGHT(2023) Booklist, ALL RIGHTS RESERVED.