S 5576: Artificial Intelligence Risk Management and Security Act of 2026

S 5576 in plain English: This bill would establish risk management and security requirements for artificial intelligence systems. It includes civil penalties of up to $250,000 per day for each continuing violation of its requirements.

Stated purpose

This bill aims to establish an Artificial Intelligence Safety Board to identify and manage risks posed by advanced AI systems, particularly those that could threaten national security, public health, or critical infrastructure.

Key points

• Establishes risk management and security standards for artificial intelligence systems
• Imposes civil penalties of up to $250,000 per violation, with each day of a continuing violation counted separately

Arguments supporters make

• Advanced AI systems are already powerful enough to potentially help bad actors create weapons or launch cyberattacks, so a dedicated safety board fills a real gap in federal oversight before a serious incident occurs.
• Having a centralized board that collects risk assessments and incident reports from AI developers gives the government a clearer, faster picture of emerging threats than the current scattered agency approach.
• Requiring developers to evaluate their own systems for dangerous capabilities and report problems creates accountability without banning AI innovation outright.

Arguments opponents make

• Defining 'frontier AI' broadly could pull a wide range of companies into a costly federal reporting regime, burdening smaller developers and startups that pose little actual risk.
• A new government board may move too slowly to keep pace with rapidly evolving AI technology, creating compliance overhead without meaningfully improving safety outcomes.
• Mandatory disclosure of sensitive capability assessments to a federal body raises concerns about government overreach and the potential exposure of proprietary information that companies need to protect competitively.

Tradeoffs

The bill trades some degree of developer privacy and reduced regulatory burden for greater government visibility into AI risks — the tension is between enabling fast-moving AI innovation and ensuring early warning of dangers that could affect public safety or national security.

Current status in Congress: In committee.

NewsClear — neutral news & congressional tracking · Bill of the Week