AI in HR: What to Expect
As legislative bodies consider new regulations regarding artificial intelligence (AI), it is important for employers to remain informed about the evolving legal landscape. Several pending bills in the United States Congress, along with anticipated legislation in the United Kingdom and the European Union, aim to regulate AI technology and provide a framework for governance.
Current Legislative Landscape – United States
In the U.S., there is currently no federal law governing the use of AI in the workplace. However, several bills are under consideration that could change this:
- No Robot Bosses Act: This bill, proposed in the summer of 2023, would prohibit employers from making employment decisions based solely on automated decision systems. Instead, employers would need to ensure human oversight of AI outputs before utilizing them in employment decisions. Additionally, the bill mandates pre-deployment auditing of AI tools for bias, periodic testing for bias, and employee training on proper AI use.
- Stop Spying Bosses Act: Although primarily focused on workplace surveillance, this bill includes provisions that restrict the use of automated decision systems for predicting employee behavior that is unrelated to their work duties.
While it is uncertain whether these bills will become law, they reflect a growing trend towards stricter regulations on AI in the workplace.
State-Level Developments
At the state level, various jurisdictions are also proposing laws aimed at limiting AI usage. For instance:
- In Illinois, HB 5116 seeks to mandate impact assessment audits for automated decision tools.
- Massachusetts is considering the Preventing a Dystopian Work Environment bill, which would require similar assessments and a list of AI tools used to be submitted to the state.
Employers should expect continued legislative activity in this area as the year progresses.
Anticipated Changes in the European Union and the United Kingdom
The U.K. is expected to introduce draft legislation on AI regulation in the coming year. The government aims to establish a legal framework that regulates AI development, moving away from the previous government’s non-binding principles. While no specific bills have been introduced yet, this shift indicates a commitment to enhancing oversight on AI.
In the European Union, the EU AI Act is already in force, with various obligations set to take effect in the coming years:
- Prohibited AI systems and AI literacy requirements will be applicable as of February 2, 2025.
- General-purpose AI model obligations will become applicable on August 2, 2025.
- High-risk AI systems will be subject to obligations starting August 2, 2026.
- Remaining provisions will be applicable by August 2, 2027.
The EU AI Act also requires guidance from various EU authorities to help organizations comply with the regulations. Additionally, data protection authorities across the EU will continue to issue guidance on AI usage.
Conclusion
As AI technology continues to evolve, it is crucial for employers to stay informed about the current and anticipated regulations. This includes understanding the implications of proposed legislation and preparing for compliance with emerging legal requirements.
