AI Bias in Hiring and Beyond: Why States Are Stepping Up Regulation

AI is already screening job resumes, rental apartment applications, and even determining medical care with almost no oversight Uncover the impact of AI bias in hiring and beyond. Learn how bias affects decision-making and explore solutions for fairer outcomes.. Studies and lawsuits have found AI systems can discriminate based on race, gender, or other factors.  Lawmakers in at least seven states are taking big legislative swings to regulate bias in artificial intelligence (AI: https://en.wikipedia.org/wiki/Artificial_intelligence, filling a void left by Congress’ inaction.

AI’s Hidden Reach

While artificial intelligence made headlines with ChatGPT: https://chat.openai.com/, behind the scenes, the technology has quietly pervaded everyday life.  It’s estimated as many as 3% of employers use algorithms to help in hiring Equal Employment Opportunity Commission https://www.eeoc.gov/

The Risks of AI Discrimination

An AI can learn bias through the data it’s trained on.  Amazon’s failed hiring algorithm: [https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK0G] is a prime example. Even with seemingly neutral factors, discrimination can still occur, as seen in a class-action lawsuit alleging bias in a rental application AI  ([Webber lawsuit – rental AI] (if you can find a  news article, link here)).

States Fill the Regulatory Void

Lawmakers are working to create regulations, but it’s a complex problem. Industry giants worth hundreds of billions of dollars will have strong opinions.  Last year, only about a dozen of the nearly 200 AI-related bills introduced in statehouses were passed into law (BSA The Software Alliance: https://www.bsa.org/

Success Depends on Complex Solutions

States are already behind in regulating AI, and its use in consequential decision-making is largely hidden.  These seven proposals focus on regulating AI across industries:

What the AI Bills Target

 Impact Assessments:  Companies would describe how AI is used, data collected, risks of discrimination, and safeguards employed.  

 Consumer Notification:  Customers would be told when an AI is involved in a decision and have the option to opt out.

Industry Pushback and the Road Ahead

Companies generally favor some regulations around impact assessments but oppose requirements to routinely test AI systems for bias,  due to concerns about revealing trade secrets. Still, this is a groundbreaking start toward regulating this ever-present technology.

Call to Action:

Are you aware of how AI is impacting your life?  Learn more about the current proposalsin California, Colorado, Rhode Island, Illinois, Connecticut, Virginia, and Vermont.