Title: The Impact of NYC’s Anti-Bias Law on Hiring Algorithms
In recent years, the use of artificial intelligence (AI) algorithms in the hiring process has become increasingly prevalent. However, concerns about bias and discrimination in these algorithms have led to the implementation of new regulations in order to ensure fair and equitable hiring practices. One of these regulations is the anti-bias law recently passed in New York City (NYC), which requires employers to submit their hiring algorithms for independent audits and disclose the results to job candidates. In this article, we will explore the potential implications of this groundbreaking law and how it may shape the future of hiring algorithms.
New York City’s Local Law 144, also known as the anti-bias law, was recently enacted to address potential biases in hiring algorithms. Employers who use these algorithms are now obligated to have them audited by an independent party and publish the findings. The law also requires employers to provide disclosures about the use of algorithms to both their staff and job applicants.
Under this law, public reports must include information about the algorithms being used and their impact on candidates of different races, ethnicities, and genders. This includes an average score given to candidates from various groups and impact ratios that compare scores across different categories. By making this information available, the law aims to shed light on any biases or disparities in hiring outcomes.
The existence of bias in various hiring algorithms has been extensively studied and documented. For example, in 2018, Amazon discontinued a recruiting engine that showed bias against female job applicants. Similarly, a 2019 academic study found AI-enabled anti-Black bias in recruiting practices. These cases illustrate the urgent need to address bias in AI algorithms and ensure fair and equitable hiring practices.
It’s important to note that the application of Local Law 144 extends beyond workers based in NYC. Anyone working or seeking employment in the city is protected by this law. Consequently, no individual, regardless of their place of birth or residence, should be subjected to discriminatory algorithms.
The New York City Department of Consumer and Worker Protection is responsible for enforcing compliance with the law, and failure to comply can result in significant fines. Companies found in violation may face fines ranging from $375 for a first offense to $1,500 for subsequent violations. Each day a company continues to use non-compliant algorithms constitutes a separate violation.
The implementation of Local Law 144 is expected to have a significant impact on the use of hiring algorithms. Despite the fact that almost a quarter of companies currently employ AI in the candidate selection process, this law introduces a new level of accountability and transparency. Businesses that fail to ensure the fairness and impartiality of their algorithms risk penalties.
There is a wide range of algorithms employed in the hiring process. These include text analyzers that classify resumes based on keywords, chatbots that conduct online interviews, and software that predicts a candidate’s problem-solving abilities and cultural fit. To comply with the law following the adoption of Local Law 144, these algorithms must undergo careful analysis and be modified where necessary.
The passage of Local Law 144 is just the first step in eliminating bias from hiring algorithms. The industry as a whole recognizes the need for self-regulation and has begun taking preliminary steps to address algorithmic bias. For example, the Data & Trust Alliance was established in December 2021 to create an evaluation and scoring system for identifying and reducing bias in AI algorithms used in hiring.
While some organizations believe Local Law 144 does not go far enough, the law does require independent audits by organizations not involved in algorithm creation or distribution. This ensures an objective and credible auditing process. However, some argue that businesses should not solely be responsible for certifying their own AI products to avoid potential conflicts of interest.
The adoption of Local Law 144 in NYC sets an important precedent for other jurisdictions considering similar regulations. It is likely to influence future legislation and industry practices. For instance, the District of Columbia is considering a rule that would hold employers legally accountable for ensuring bias-free automated decision-making systems. California has also proposed legislation to regulate the use of AI in hiring, and New Jersey recently introduced a bill to reduce the likelihood of discrimination in AI-based hiring decisions.
These legislative efforts highlight the growing recognition of the need to address bias in AI algorithms, particularly in the context of hiring. As technology advances, it is crucial to ensure that these developments do not perpetuate existing prejudices or perpetuate discrimination in the workplace.
The passage of the anti-discrimination law in New York City is a significant step towards achieving fair and transparent hiring practices. The law aims to identify and rectify potential biases in algorithms through independent audits and public disclosures. While its effectiveness remains to be seen, it will likely shape future legislation and industry practices. Businesses must prioritize fairness and equity as they integrate AI into their recruitment processes, fostering a more inclusive and diverse workforce.
