In Washington D.C., a proposed bill, titled the "Stop Discrimination by Algorithms Act", faces criticism for potentially infringing upon the fair treatment of algorithms, while aiming to stop discrimination against individuals.
The Stop Discrimination by Algorithms Act, proposed in Washington D.C., aims to prevent discriminatory outcomes caused by automated decision-making systems. Although the exact bill text is not available, the Act is expected to include several key provisions based on AI discrimination legislation and regulatory trends.
The Act applies to organizations and entities that use automated decision systems (algorithms) in consequential decisions, such as employment, housing, credit, or public accommodations. This includes both private companies and government agencies operating in D.C.
The Act prohibits the use of algorithms that result in discrimination based on protected characteristics such as race, sex, disability, age, and other legally protected classes. This includes both intentional discrimination and disparate impact—where a seemingly neutral algorithm disproportionately harms a protected group.
Disclosure is a crucial aspect of the Act. Organizations must disclose when automated decision-making tools are used, especially in contexts affecting individuals’ rights or access to services. This is intended to ensure transparency, enabling individuals to understand when AI influences decisions about them.
Regular bias audits or impact assessments of algorithms are required under the Act. These audits help detect and mitigate discriminatory effects and ensure compliance with anti-discrimination laws. Audits may need to be reported to regulatory authorities or made available for inspection.
The Act also establishes enforcement powers for the D.C. Attorney General to investigate violations, issue penalties or fines, and require remedial actions from violators. Enforcement can include responding to complaints, conducting audits, and imposing sanctions.
The private right of action in the Act allows individuals to bring a civil suit against organizations that violate the Act. However, concerns have been raised that the law could result in more consumers being denied access to "important life opportunities" due to less accurate decisions and reduced human bias reduction in decision-making.
The proposed Act could discourage many organizations from using AI, especially those serving D.C. residents, due to additional compliance burdens and increased liability. Additionally, the floodgate of frivolous lawsuits created by the private right of action in the Act could impose substantial costs on organizations, which may be passed on to consumers.
For the most recent bill drafts or press releases, it would be advisable to consult Washington D.C.’s government or council websites. The Act reflects the growing legal focus on preventing algorithmic discrimination by mandating transparency, accountability, and enforcement to protect individuals against unfair AI-driven outcomes.
- The Stop Discrimination by Algorithms Act proposes to regulate automated decision-making systems in digital economy entities operating in Washington D.C., covering both private companies and government agencies.
- The Act prohibits the use of algorithms that result in discrimination based on protected characteristics, such as race, sex, disability, age, and other legally protected classes, including both intentional and disparate impact.
- Transparency is a cornerstone of the Act, requiring organizations to disclose when automated decision-making tools are used, particularly in contexts affecting individuals’ rights or access to services.
- Regular bias audits or impact assessments of algorithms are mandated under the Act to ensure compliance with anti-discrimination laws and detect/mitigate discriminatory effects.
- The D.C. Attorney General is granted enforcement powers to investigate violations, issue penalties or fines, and require remedial actions from offenders under the Act.
- The Act allows individuals to bring civil suits against organizations that violate the Act, advocating for fair AI-driven outcomes in the areas of employment, housing, credit, and public accommodations, but concerns have been raised about potential negative impacts on AI usage and consumer access.
