President Joe Biden recently signed an
Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence aimed at regulating artificial intelligence (AI) by establishing standards for security, privacy protections, and safety testing of new AI models. The order emphasizes governance of AI technology to leverage its potential while averting risks associated with its rapid development and public usage.
Key points of the executive order include:
1. Safety testing and standards: Companies developing powerful AI tools, including major entities such as Microsoft, Amazon, and Alphabet’s Google, are required to submit test results of their new AI models to the government for assessment before releasing them to the public.
2. Watermarking AI-Generated content:: The Commerce Department will establish standards for watermarking AI-generated content like audio or images, commonly known as “deepfakes,” enabling verification of authenticity to distinguish between genuine and altered content.
3. Government’s role in vetting technology: The US government, being a significant customer for tech companies, will vet technologies with potential national security threats or economic and health risks.
4. Federal Agency Involvement: The order will involve multiple federal agencies. The Department of Energy will ensure AI systems don’t pose chemical, biological, or nuclear risks, while the Departments of Defense and Homeland Security will focus on developing cyber protections for safer critical infrastructure and computers.
5. Legislative action on privacy concerns: Biden plans to meet with Senate Majority Leader Chuck Schumer and lawmakers to discuss passing legislation addressing privacy concerns associated with AI technology.
6. Global collaboration on AI risks:: Vice President Kamala Harris expressed the need for global collaboration to address the risks of technology with worldwide implications. This aligns with the US plan to collaborate with allies and apply international rules and norms to promote global order and stability.
The executive order builds upon voluntary commitments from several companies earlier in the year and aligns with the proposed “AI Bill of Rights” focusing on safe AI development and use.
Additionally, the order addresses concerns regarding algorithmic bias, instructing the Justice Department to set best practices for investigating and prosecuting civil rights violations related to AI, particularly in housing and the criminal justice system. It also aims to ease visa requirements for overseas talent seeking to work at American AI companies.
The order is a significant step by the US government to govern AI technology and balance its growth while safeguarding against potential risks and ensuring responsible AI deployment.
