The article argues that employment law is increasingly moving upstream, regulating not only what employers do when making hiring or employment decisions, but also the information that reaches them in the first place. Artificial intelligence, criminal records, consumer credit reports, privacy, and automated decision-making may appear to be separate issues, but they share a common theme: lawmakers are increasingly examining how employment-related information is created, maintained, accessed, evaluated, and processed before it influences a decision.
The article gives several examples of this trend. Illinois' Clean Slate laws address the accuracy and availability of criminal records, while Missouri has introduced automated expungement for certain qualifying offenses. New York has restricted employers' use of consumer credit information, partly because of concerns about its relevance to job performance and potential inaccuracies. Washington's Fair Chance Act goes further by requiring employers to consider context and mitigating circumstances before taking adverse action based on certain criminal records.
Artificial intelligence is becoming the latest area where these principles are being applied. The article discusses proposed California legislation that would place restrictions on certain automated employment-related inferences and require human corroboration for specified AI-generated outputs before they can support consequential decisions. The underlying concern is that an automated system can produce information that is inaccurate, biased, misleading, or difficult to explain. As employers increasingly rely on HR technology and AI tools, the reliability and provenance of the information behind an employment decision become legally important.
The broader message is that the future of employment regulation may focus increasingly on the entire information lifecycle, rather than only on the final hiring or firing decision. Employers may need to understand where their data comes from, whether it is accurate, what information should or should not be considered, and how automated systems process it. In this sense, AI regulation, privacy rules, fair-chance laws, and reporting requirements can be viewed as different ways of controlling the information that ultimately shapes employment decisions.
