U.S. Releases Strategic Vision on AI Safety, Plan for Global Cooperation

Author photo: Patrick Arnold
ByPatrick Arnold
Company and Product News

As the AI Summit in Seoul, South Korea begins, the United States has released a strategic vision for the U.S. Artificial Intelligence Safety Institute (AISI), describing the department’s approach to AI safety. The National Institute of Standards and Technology (NIST) within the Department of Commerce launched the AISI, building on the organization’s long-standing work on AI. In addition to releasing a strategic vision, the department shared plans to work with a global scientific network of AI Safety Institutes and other government-backed scientific offices, and to convene the institutes later this year in San Francisco.

Vision on AI Safety,

Advancements in AI technology represent huge potential for society but pose real danger if not developed and deployed in a responsible way. The U.S. AI Safety Institute engages with academia, industry, and the public sector to better evaluate and mitigate the risks of AI, with the fundamental goal of harnessing the benefits. The strategic vision outlines how to achieve that objective through cooperation with a global scientific network on AI safety.

The AISI will focus on three key goals:

  • Advance the science of AI safety

  • Articulate, demonstrate, and disseminate the practices of AI safety

  • Support institutions, communities, and coordination around AI safety

To achieve these goals, the AISI plans to, among other activities, conduct testing of advanced models and systems to assess potential and emerging risks; develop guidelines on evaluations and risk mitigations, among other topics; and perform and coordinate technical research.

Concurrently, the AISI will help launch a global scientific network for AI safety through meaningful engagement with AI Safety Institutes and other government-backed scientific offices focused on AI safety. This network will strengthen and expand on AISI's previously announced collaborations with the AI Safety Institutes of the UK, Japan, Canada, and Singapore, as well as the European AI Office and its scientific components and affiliates.

See the full text of the strategic vision document here.

For more ARC coverage on AI and other topics, please visit our website.

Engage with ARC Advisory Group

Representative End User Clients
Representative Automation Clients
Representative Software Clients