The Implications of Artificial Intelligence on Civil Rights
Artificial intelligence (AI) has become deeply integrated into various aspects of our lives, from mortgage lending algorithms to hiring tools and automatic systems used in hospitals. However, as technological advancements continue, it is important to assess how AI can potentially impact civil rights and perpetuate discrimination.
Recently, Sen. Cory Booker (D-N.J.) gathered a panel of experts and advocates to discuss the implications of AI on civil rights. He acknowledged the power of technology and innovation but also emphasized that technology has the potential to amplify discrimination. This sentiment was echoed by many panelists who highlighted specific instances where AI has disproportionately affected marginalized communities.
One notable example discussed was mortgage lending algorithms, which were found to be more likely to deny home loans to people of color compared to white individuals. Similarly, AI-enabled recruiting and hiring tools have been known to discriminate against women and minority candidates. Moreover, automatic systems used in hospitals have repeatedly failed to accurately assess the needs of Black patients, exacerbating healthcare disparities.
Additionally, the panel highlighted instances where AI has altered people’s physical appearances, including lightening skin tones. They also pointed out chatbots like ChatGPT that have produced stereotypes and even slurs. It is evident that AI can perpetuate harmful biases and stereotypes, underscoring the urgent need to address these issues.
The panelists emphasized that equal opportunity, civil rights, and racial justice are deeply intertwined with today’s and tomorrow’s technology. Algorithms are extensively used to make decisions in various aspects of our lives, ranging from determining bail to housing and education accessibility. However, these algorithms often rely on historical data that may reflect generations of redlining, segregation, and discrimination. As a result, they are prone to perpetuating harmful biases and harming marginalized communities.
Nevertheless, the panel also explored how AI can be used to counter these issues. Suresh Venkatasubramanian, director of the Center for Technological Responsibility, Reimagination, and Redesign at Brown University, emphasized that humans have the power to eliminate racial biases in the construction of AI systems. AI systems are designed to do what humans instruct them to do, so by testing, examining the data used to train the system, and making fair design choices, discriminatory outcomes can be minimized.
Some panelists, like community advocate Fabian Rogers, highlighted the necessity for legislative protections, especially concerning facial recognition software. The American Civil Liberties Union has previously raised concerns about the potential misuse of facial recognition technology to track individuals at protests and religious gatherings. These advocates called for the passage of an AI Bill of Rights to establish clear guidelines and protect individuals from the harmful effects of AI.
Fortunately, steps are already being taken to address these concerns. The White House has released a five-point blueprint that identifies principles to guide the design, use, and deployment of AI systems to protect Americans. These principles include consultation with diverse communities, equitable design to limit artificial bias, data privacy protections, transparency regarding AI use, and the right to opt out of AI systems.
Furthermore, Congress is currently working on passing legislation to regulate the use of AI and protect individuals affected by its decisions. Sen. Cory Booker, Sen. Ron Wyden (D-Ore.), and Rep. Yvette Clarke (D-N.Y.) recently introduced the Algorithmic Accountability Act of 2023. This bill aims to create new protections for individuals impacted by AI systems in areas such as housing, credit, and education.
The discussions surrounding the implications of AI on civil rights highlight the importance of ethical and responsible development of AI systems. While AI has the potential to perpetuate discrimination, it also holds the power to identify and rectify biases. By ensuring fair design choices, thorough testing, and implementing legislation to protect individuals, we can harness the potential of AI to create a more equitable and just society.