Rep. Ro Khanna will introduce the Human Control Over AI Act, establishing strict liability standards and banning recursive self-improving AI systems until government safeguards are in place.
The Silicon Valley Democrat's proposed legislation targets the development and deployment of advanced AI systems, requiring companies to maintain meaningful human oversight and control.
The bill's key provisions include:
Strict Liability Framework: Companies developing or deploying AI systems would face legal responsibility for harms caused by their technology. This shifts accountability directly to organizations rather than treating AI as an autonomous entity.
Self-Improvement Ban: The legislation prohibits recursive self-improving AI—systems that autonomously enhance their own capabilities—until the government establishes adequate safeguards. This creates a regulatory checkpoint before such systems advance further.
Human Control Requirement: The act prioritizes maintaining human decision-making authority over AI operations, preventing systems from operating without meaningful human intervention in critical contexts.
Khanna's proposal joins growing Congressional scrutiny of AI regulation. As artificial intelligence systems become more capable and widely deployed, lawmakers have increasingly focused on preventing potential harms while maintaining innovation.
The bill reflects concerns about AI systems operating without adequate oversight, particularly regarding advanced models that could operate with minimal human intervention. By establishing liability standards, the legislation aims to incentivize safety measures during development and deployment.
The self-improvement ban targets a specific concern among AI researchers and policymakers: systems that modify their own code or structure without human review could exceed intended capabilities and become difficult to control.
The introduction of this bill signals continued legislative momentum around AI governance. Other proposals in Congress have focused on transparency requirements, algorithmic accountability, and safeguards for specific applications like facial recognition and hiring tools.
The Human Control Over AI Act will likely face debate over the feasibility of strict liability standards, the scope of systems affected, and whether the measures adequately protect public interests without stifling AI development.
As artificial intelligence increasingly tackles complex mathematical problems through brute computational force, mathematicians are grappling with how to preserve the creative and artistic dimensions that have defined their field for centuries.
A Wuhan court has included token usage and AI tool licensing fees in a copyright infringement damages calculation—a first for China's legal system. The ruling reflects broader efforts to establish copyright protections for AI-generated works.
Artificial intelligence and data centers overshadowed traditional climate solutions at Climate Week, creating tension between tech entrepreneurs and investors who disagree on AI's role in addressing environmental challenges.
OpenAI is attempting to mend relations with mathematicians after repeated missteps in announcing breakthroughs. The company has formed an independent advisory group of elite math practitioners to improve communication.