President Joe Biden has signed an expansive executive order on artificial intelligence, encompassing various aspects ranging from civil rights and industry regulations to a significant increase in government involvement and investment in AI technology.
In a media briefing prior to the order’s signing, a senior White House official emphasized the multifaceted nature of AI policy, comparing it to a decathlon with ten different events. The official stressed that effective regulations must cover a broad spectrum of concerns, including safety, equity, and privacy, making it imperative to address all these aspects comprehensively.
Furthermore, the official called for “significant bipartisan legislation” to advance the nation’s interests in AI. Despite a private forum convened by Senate Majority Leader Chuck Schumer with industry leaders in September, substantial AI legislation has yet to be introduced.
The executive order builds upon a nonbinding agreement reached in July by seven leading U.S. tech companies involved in AI development. The agreement included commitments to hire external experts to assess their AI systems for vulnerabilities and to share critical findings. The order now legally obligates these companies to share safety test results with the federal government, leveraging the Defense Production Act.
Additionally, the order tasks the Commerce Department with creating guidelines for “watermarking” AI-generated content to clearly indicate that it was not created by humans. This measure aims to address concerns related to deepfake videos and AI-generated essays.
The executive order also allocates additional funding for AI research and initiates a federal hiring surge in the AI field. To facilitate these initiatives, the White House has launched the AI.gov website, connecting job seekers with AI-related government positions.
Fei-Fei Li, co-director of Stanford’s Institute for Human-Centered Artificial Intelligence, stressed the importance of government funding in enabling AI to tackle critical societal challenges. She highlighted the potential of the public sector to leverage data and interdisciplinary talent for tasks such as curing diseases, mapping biodiversity, predicting wildfires, and addressing climate change.
Sarah Myers West, managing director of the AI Now Institute, praised President Biden for incorporating social and ethical concerns into the executive order. She emphasized that the order’s focus on labor, civil rights, privacy protection, and competition reflects the public’s top priorities.
Myers West also emphasized the need for strong enforcement to prevent companies from setting self-regulatory standards, underscoring that the industry should not lead the conversation on addressing the effects of AI on the broader public.