Businesses are screening prospective job candidates with tech-assisted cognitive and personality tests, deploying wearable technology to monitor brain activity on the job and using artificial intelligence to make decisions about hiring, promoting and firing people. The brain is becoming the ultimate workplace sorting hat — the technological version of the magical device that distributes young wizards among Hogwarts houses in the “Harry Potter” series.
Companies touting technological tools to assess applicants’ brains promise to dramatically “increase your quality of hires” by measuring the “basic building blocks of the way we think and act.” They claim their tools can even decrease bias in hiring by “relying solely on cognitive ability.”
But research has shown that such assessments can lead to racial disparities that are “three to five times greater than other predictors of job performance.” When social and emotional tests are part of the battery, they may also screen out people with autism and other neurodiverse candidates. And applicants may be required to reveal their thoughts and emotions through AI-based, gamified hiring tools without fully understanding the implications of the data being collected. With recent surveys showing that more than 40% of companies use assessments of cognitive ability in hiring, federal employment regulators have rightly begun to pay attention.
Once workers are hired, new wearable devices are integrating brain assessment into workplaces worldwide for attention monitoring and productivity scoring on the job. The SmartCap tracks worker fatigue, Neurable’s Enten headphones promote focus and Emotiv’s MN8 earbuds promise to monitor “your employees’ levels of stress and attention using … proprietary machine learning algorithms” — though, the company assures, they “cannot read thoughts or feelings.”
The growing use of brain-oriented wearables in the workplace will undoubtedly put pressure on managers to use the insights gleaned from them to inform hiring and promotion decisions. We are vulnerable to the seductive allure of neuroscientific explanations for complex human phenomena and drawn to measurement even when we don’t know what we should be measuring.
Relying on AI-based cognitive and personality testing can lead to simplistic explanations of human behavior that ignore the broader social and cultural factors that shape the human experience and predict workplace success. A cognitive assessment for a software engineer may test for spatial and analytical skills but ignore the ability to collaborate with people from diverse backgrounds. The temptation is to turn human thinking and feeling into puzzle pieces that can be sorted into the right fit.
The U.S. Equal Employment Opportunity Commission seems to have awakened to these potential problems. It recently issued draft enforcement guidelines on “technology-related employment discrimination,” including the use of technology for “recruitment, selection, or production and performance management tools.”
While the commission has yet to clarify how employers can comply with nondiscrimination statutes while using technological assessments, it should work to ensure that cognitive and personality testing is limited to employment-related skills lest it intrude on the mental privacy of employees.
The growing power of these tools may tempt employers to “hack” candidates’ brains and screen them based on beliefs and biases, assuming such decisions aren’t unlawfully discriminatory because they aren’t directly based on protected characteristics. Facebook “likes” can already be used to infer sexual orientation and race with considerable accuracy. Political affiliation and religious beliefs are just as easily identifiable. As wearables and brain wellness programs begin to track mental processes over time, age-related cognitive decline will also become detectable.
All of this points to an urgent need for regulators to develop specific rules governing the use of cognitive and personality testing in the workplace. Employers should be required to obtain informed consent from candidates before they undergo cognitive and personality assessment, including clear disclosure of how candidates’ data is being collected, stored, shared and used. Regulators should also require that assessments be regularly tested for validity and reliability to ensure that they’re accurate, reproducible and related to job performance and outcomes — and not unduly sensitive to factors such as fatigue, stress, mood or medications.
Assessment tools should also be regularly audited to ensure that they don’t discriminate against candidates based on age, gender, race, ethnicity, disability, thoughts or emotions. And companies developing and administering these tests should regularly update them to account for changing contextual and cultural factors.
More broadly, we should consider whether these methods of assessing job applicants are promoting excessively reductionist views of human abilities. That’s especially true as the capabilities of human workers are more frequently compared with those of generative AI.
While the use of cognitive and personality assessments is not new, the increasing sophistication of neurotechnology and AI-based tools to decode the human brain raises important ethical and legal questions about cognitive liberty.
Employees’ minds and personalities should be subject to the most stringent protection. While these new tests may offer some benefits for employers, they must not come at the cost of workers’ privacy, dignity and freedom of thought.
Nita Farahany is a professor of law and philosophy at Duke University and the author of “The Battle for Your Brain: Defending the Right to Think Freely in the Age of Neurotechnology.”