White House AI council to meet on Monday amid legislative stagnation

Convening three months after President Biden’s executive order on AI risks, the White House Council declares new regulations to examine foreign access to U.S. cloud companies under proposed KYC rules.
Flag, Landmark, The White House, Animal, Bird

The White House AI council met on Monday, 29 January, to address the increasingly significant risks AI systems pose. The White House Deputy Chief of Staff Bruce Reed highlighted President Biden’s directive to “move fast and fix things.” In compliance with the executive order, nine US government agencies, including defense, transportation, treasury, and health, have already provided risk assessments to the Department of Homeland Security.

Notably, on Friday, the Biden administration put forward a significant proposal to compelling US cloud companies, including Amazon’s AWS, Alphabet’s Google Cloud, and Microsoft’s Azure unit, to follow “know your customer” rules for cloud services. The aim is to adequately scrutinize whether foreign entities are accessing US data centers for AI model training. US Commerce Secretary Gina Raimondo pushed for the proposal, stating the importance of ensuring that non-state actors or China cannot access US data centers. 

Raimondo further revealed that under the Defence Production Act, the Commerce Department aims to make it mandatory for AI system developers working in critical industries to share their AI safety test findings with the government before publicly releasing their AI systems. The companies will soon be sent survey requests, where non-compliance within the stipulated 30 days by a company will be viewed as a matter of concern.

This move comes from the increasing focus on the development of responsible AI, confirming the growing unease around the potential risks and challenges associated with the technology.