President Trump signs executive order to strengthen cyber defenses while rejecting AI regulations.
AI's cyberattack capabilities are increasing year by year, and AI company Anthropic has voluntarily refrained from publicly releasing its '
Promoting Advanced Artificial Intelligence Innovation and Security – The White House
https://www.whitehouse.gov/presidential-actions/2026/06/promoting-advanced-artificial-intelligence-innovation-and-security/
The executive order signed by President Trump on June 2, 2026, aims to encourage AI innovation within the government while also preparing the government to assess the safety of AI.
Specifically, the Department of War (Department of Defense) was ordered to prioritize strengthening the cyber defense capabilities of the national security system. Other agencies were instructed to create a system to accept when AI developers voluntarily share security information with the government, and to define advanced models as 'target frontier models,' properly evaluate them, and share the results with AI developers and researchers.
Furthermore, the presidential decree explicitly states that 'these provisions should not be interpreted as requiring government permission for AI development,' thus continuing to encourage free AI development.
Executive Order No. 1 states that 'America continues to lead the world in the field of AI because it refuses to stifle innovation with overly burdensome regulations. My administration has achieved remarkable technological growth and economic investment in the field of AI by removing the bureaucratic constraints imposed by the previous administration on AI developers and researchers and instead encouraging AI innovation.' It also states that while maintaining this system, the government needs to strengthen its defense capabilities to deal with impending crises.
Under these new regulations, AI developers can consult with the federal government to determine whether their model under development meets the criteria for a 'Target Frontier Model,' and, subject to compliance with confidentiality, cybersecurity, internal risk, intellectual property protection, usage, and secrecy requirements, they can provide the Target Frontier Model to the federal government up to 30 days before its planned release date. They can also collaborate with the federal government to select partners who will be granted early access to the Target Frontier Model.
Leading AI company Anthropic's relationship with the U.S. government deteriorated over disagreements regarding AI, but in April 2026, CEO Dalio Amodi attended a White House meeting where 'fruitful discussions took place on how Anthropic and the government can cooperate.' Technology media outlet The Verge reported that 'the Mythos model appears to have been the key to easing tensions with the government.'
Related Posts:
in AI, Posted by log1p_kr