The U.S. Department of Labor released a list of artificial intelligence best practices for developers and employers this week. Here’s the PDF.
The highlights:
- Human-Centric AI Development: The document emphasizes that advancements in AI should prioritize human agency and creativity, aiming to enhance worker well-being and improve job quality.
- Worker Engagement: It advocates for the active involvement of workers, especially from underserved communities, in all stages of AI development and deployment to ensure their needs and rights are considered.
- Ethical Standards: Developers and employers are encouraged to establish ethical standards for AI systems that protect workers’ rights, mitigate risks, and ensure safety, thereby enhancing overall job quality.
- Governance and Oversight: Organizations should implement clear governance structures and human oversight for AI systems to ensure accountability and to mitigate risks associated with their use.
- Transparency in AI Use: Employers must provide workers with clear information regarding the AI systems in use, their purpose, and the data collected, fostering trust and security in the workplace.
- Protection of Worker Rights: The document stresses that AI systems should not undermine workers’ rights to organize, health and safety, or anti-discrimination protections, ensuring compliance with legal obligations.
- Job Quality Enhancement: AI should be deployed in ways that assist and complement workers, improving job quality rather than automating away good jobs, thus maximizing benefits for both employees and employers.
- Support for Displaced Workers: Employers are encouraged to provide retraining and upskilling opportunities for workers affected by AI-related job transitions, promoting a proactive approach to workforce management.
- Responsible Data Use: The document outlines the importance of limiting the collection and use of worker data to legitimate business purposes while ensuring data protection and privacy.
- Continuous Evaluation: Regular independent audits and evaluations of AI systems are recommended to ensure they are functioning as intended and to identify any adverse impacts on workers, allowing for timely corrections.