This decision follows months of internal deliberations regarding the national security and cyber defense ramifications of AI. The White House, having moved away from its prior lenient approach, has opted for increased oversight of the rapidly evolving AI landscape.
The new directive comes on the heels of the retraction of a previous executive order last month that aimed to create a review timeline of 14 to 90 days for new AI models. This earlier proposal was dropped just before it was scheduled to be signed by Trump.
Under the revised framework, companies creating advanced AI technologies will be required to submit their models for government evaluation 30 days prior to their public debut. The administration stated that this condensed review period aims to reconcile oversight needs with ongoing technological advancements.
The order also mandates the Treasury Secretary to form an AI cybersecurity clearinghouse responsible for assessing security vulnerabilities identified by AI models and coordinating responses to potential threats.
The executive order emphasizes that while state-of-the-art AI capabilities can bolster the United States, they also present national security challenges that necessitate collaborative action across government entities.
This decision marks a significant departure from the administration’s earlier standpoint, which promoted minimal regulation, arguing that fewer constraints would enable the United States to retain a competitive edge in technological rivalry against China.
