Bipartisan Framework for U.S. AI Act

• Authority: Other authorities
• Collections: Miscellaneous documents
• Most recent activity: Proposed
• Most recent activity date: 2023-09-07

Risk subdomain coverage

Good coverage — Comprehensive governance measures or mitigations with clear procedures relating to the risk subdomain are explicitly described.

• Governance failure
• Lack of transparency or interpretability

Minimal coverage — Brief mention but no specific mitigations, recommendations or governance measures relating to the risk subdomain are described.

• Unfair discrimination and misrepresentation
• Exposure to toxic content
• Compromise of privacy
• AI system security vulnerabilities
• False or misleading information
• Pollution of information ecosystem
• Disinformation, surveillance, and influence at scale
• Cyberattacks, weapon development or use
• Fraud, scams, and targeted manipulation
• Overreliance and unsafe use
• Loss of human agency and autonomy
• Increased inequality and decline in employment
• Competitive dynamics
• AI possessing dangerous capabilities
• Lack of capability or robustness

All other risk subdomains had no coverage in this document.

Risk Coverage Summary

The document has good coverage of approximately 12-14 subdomains, with strong focus on malicious actors (4.1, 4.2, 4.3), privacy compromise (2.1), AI system security (2.2), false information (3.1), governance failure (6.5), competitive dynamics (6.4), lack of transparency (7.4), and lack of robustness (7.3). Coverage is concentrated in security, misuse prevention, transparency, and governance domains.

Sector coverage

Good coverage — Comprehensive governance measures or mitigations with clear procedures relating to the sector are explicitly described.

• Information
• National Security

Minimal coverage — Brief mention but no specific mitigations, recommendations or governance measures relating to the sector are described.

• Scientific Research & Development Services
• Public Administration (excl. National Security)

All other sectors had no coverage in this document.

Sectors Summary

This is an external regulatory framework that applies broadly across multiple sectors. It explicitly governs AI use in high-risk situations including facial recognition, and addresses sector-specific concerns in national security, information/technology, and implicitly covers any sector deploying AI in consequential decision-making contexts. The framework is cross-sectoral in nature.

Legislative Status

Other

Legislative Status Summary

This is a proposed legislative framework that has not yet been enacted into law. It uses prescriptive language ('should be required') indicating intended future hard law, but currently represents a policy proposal or blueprint rather than binding legislation.

Actors

Proposer

• Entity Types: AI_Governance_Actor
• Summary: The document is titled 'Bipartisan Framework for U.S. AI Act' and consistently addresses recommendations to Congress, suggesting it was proposed by legislators or policy groups to Congress.

Target

• Entity Types: AI_Developer, AI_Deployer
• Summary: The framework explicitly targets both AI developers (those creating models) and AI deployers (those using AI systems in various contexts), with specific requirements for each group.

Enforcer

• Entity Types: AI_Governance_Actor
• Summary: The framework proposes creation of an independent oversight body with enforcement authority, including auditing powers and cooperation with state Attorneys General who would have concurrent enforcement authority.

Monitor

• Entity Types: AI_Governance_Actor
• Summary: The proposed independent oversight body would have monitoring responsibilities including tracking technological developments, economic impacts, maintaining a public database, and reporting on AI incidents and harms.

AI Lifecycle Stages

Lifecycle stages covered by the documents, as defined in the OECD Framework for the Classification of AI Systems

• Plan and Design
• Collect and Process Data
• Build and Use Model
• Verify and Validate
• Deploy
• Operate and Monitor

AI Lifecycle Summary

The document covers multiple lifecycle stages with primary focus on deployment, operation and monitoring. It addresses pre-deployment testing and risk management (Build and Use Model, Verify and Validate), deployment requirements including registration and licensing (Deploy), and ongoing monitoring and incident reporting (Operate and Monitor). Planning and design stage is implicitly covered through risk management requirements.

Technical Scope

Terms relating to technical scope that are mentioned or implicitly referenced in the document

• AI Models
• AI Systems
• General Purpose AI
• Generative AI

Technical Scope Summary

The document explicitly mentions AI models and AI systems throughout. It specifically references general-purpose AI models (e.g., GPT-4) and generative AI. It does not explicitly mention frontier AI, foundation models, task-specific AI, predictive AI, open-weight models, or specific compute thresholds.