Experts suggest that Senator Mitt Romney's call for increased federal oversight of artificial intelligence (AI) faces challenges due to the rapid advancement and wide-ranging applications of the technology. While many agree that AI poses risks, determining which systems need strict monitoring proves difficult. Finding a balance between encouraging technological developments and managing the associated risks is crucial. Existing safeguards for large language models (LLMs) have proven vulnerable to exploitation. Currently, the US has limited federal legislation addressing AI, while the European Parliament has passed the pioneering Artificial Intelligence Act to regulate data quality, transparency, human oversight, and accountability. Romney, along with other senators, has proposed four ways to regulate AI, including creating a new committee and agency for oversight.

Implementing these measures would require clearly defining which types of AI and use cases would fall under regulation, and potentially passing new laws to grant special powers to new government agencies. Experts emphasize the need for a thoughtful balance that prevents hindrance to technological advancement without compromising safety. Additionally, global competition in AI development should be considered. While opinions on the threat AI poses vary among experts, concerns mainly center around adversaries using AI to enhance cyber-attacks. Despite the potential risks, AI offers numerous benefits but needs robust regulatory frameworks to preemptively mitigate misuse.