White House Unveils AI Action Plan, Prioritizes Deregulation Over Ethics
The White House introduced its “AI Action Plan” on July 23, 2023, aiming to accelerate artificial intelligence innovation through deregulation and privatization of infrastructure. This 24-page document emphasizes a competitive approach to AI, framing it as a race rather than a technology that requires governance and ethical considerations.
Key elements of the plan include directives from former President Donald Trump, who signed three executive orders to expedite federal approval for data centers and to promote the export of AI models to maintain American dominance. The plan notably fails to address ethical concerns, mentioning “responsibility” only once in relation to securing AI from adversarial attacks.
While the action plan contains a section titled “Build World-Class Scientific Datasets,” which acknowledges human rights, it lacks concrete measures for protecting vulnerable populations. The plan does not address critical ethical issues such as the treatment of children, neurodivergent individuals, and minorities, which are essential considerations highlighted in the European Union AI Act.
In a concerning move, the plan suggests a revision of the NIST AI Risk Management Framework to remove references to misinformation, diversity, equity, inclusion, and climate change. This approach implies that these significant issues may be marginalized or censored, potentially affecting how AI systems interact with information on these topics.
The absence of oversight mechanisms raises significant accountability concerns. By rejecting what it deems “onerous regulation,” the administration is effectively endorsing opaque AI systems. This lack of transparency could lead to negative outcomes, such as discrimination in hiring practices and healthcare, without clear accountability for the results produced by AI systems.
The plan delegates oversight responsibilities to private corporations, relying on self-regulation instead of comprehensive governance. This hands-off strategy reflects a broader trend in deregulation, as noted during a Senate hearing led by U.S. Senator Ted Cruz, where a “light-touch regulatory style” was praised as an effective governance strategy.
Additionally, the plan creates a double standard regarding data sharing. It mandates that federally funded researchers disclose non-proprietary datasets used in AI research while allowing private companies to retain proprietary data without similar obligations. This imbalance risks reinforcing the dominance of large tech companies, as public research efforts may inadvertently support private profits.
Moreover, the plan’s disregard for copyright issues raises concerns about the ethical implications of data scraping. This could lead to a culture where data is extracted without proper attribution, disincentivizing researchers from sharing valuable information openly. The rationale provided by Trump during a White House AI summit suggested that successful AI development should not require payment for every piece of knowledge utilized, despite previous agreements between AI companies and publishers aimed at establishing fair licensing practices.
While the AI Action Plan could potentially enhance corporate AI initiatives in the United States, it raises questions about the erosion of democratic values that the country has long championed. By prioritizing national self-interest in the AI landscape, there is a risk that the U.S. might achieve technological supremacy at the cost of the ethical standards it professes to uphold.
Mohammad Hosseini, Ph.D., an assistant professor at Northwestern University’s Feinberg School of Medicine, emphasizes the importance of addressing these ethical challenges. As the global conversation around artificial intelligence continues to evolve, the implications of this action plan will likely resonate far beyond American borders.
