Job & Immigration Applicants - AI Bias Lawsuits & Algorithmic Harm

Facing large volumes of applications and processing backlogs, companies and governments are turning to AI to speed life-changing employment and immigration decisions. Frequent AI biases and errors have led to lawsuits. This month's edition of The Quantum Record looks at the rapidly increasing use of AI for filtering job and immigration applicants and conducting employment interviews, and how people who have suffered algorithmic injustice are turning to the courts and demanding regulations for protection. Read the article "Biases and Flaws of AI Decision-Making for Employment and Immigration are Upending Lives" at https://lnkd.in/ePRh7ZFw. California-based Eightfold AI Inc. is facing a potential class action suit for discriminatory practices with products like its AI Interviewer, which the company’s website claims can “run 1 million interviews in 1 hour.” The decision-making criteria applied by the company's algorithms are hidden from job applicants, who have no ability to address errors and misinterpretations. A class action suit is proceeding against the company Workday Inc. for discrimination by its job applicant screening algorithms. The company’s motion for dismissal was rejected by the judge based on evidence that included a rejection notice received by one job seeker at 1:50 a.m., less than an hour after submitting his application. Although safeguards might require a human in the loop for final decision-making, they can easily fail when the human reviewer lacks the time or training to vet all of the data in an AI-generated recommendation. With a backlog of nearly 1 million immigration applications, the Canadian Immigration Department's human in the loop requirement clearly failed to prevent a recent AI-generated rejection of an immunology PhD's application, even though the AI had hallucinated the applicant's background as a software coder. What's the solution? Is there a way to strike a balance between speed and justice? Have you faced algorithmic injustices of these types? Drop me a line, I'm interested to hear your story.