AI Governance Institute
← News
Research2026-05-26

AI Governance Problems, Policy Options, and Research Gaps Mapped in LawAI Literature Review

What happened

LawAI published the Advanced AI Governance: A Literature Review of Problems, Options and Research Challenges in January 2025, providing a structured survey of academic and policy literature on frontier AI governance mechanisms. The review covers regulatory instruments including compute security measures, software and hardware export controls, licensing regimes for advanced AI systems, structured system evaluations, and procurement rules designed to advance AI safety objectives. It also examines voluntary corporate governance proposals such as Responsible Scaling Policies adopted by several leading AI developers, and formal AI certification schemes being explored by standards bodies and regulators. The document does not make binding recommendations but synthesizes existing research to identify where evidence is strong, where significant gaps remain, and which governance questions require further empirical or legal investigation. The review is intended to serve policymakers and governance practitioners across international frameworks, with particular relevance to active policy discussions in the United States, the United Kingdom, and the European Union.

Why it matters

  • ·The review signals that licensing and pre-market certification regimes for frontier AI models are under active consideration across multiple jurisdictions, creating potential new regulatory exposure for developers who have not yet mapped their compliance posture against emerging approval obligations.
  • ·The document's detailed treatment of system evaluations connects directly to existing mandates such as the EU AI Act's conformity assessment requirements and NIST's AI Risk Management Framework, meaning organizations already subject to these frameworks may face heightened scrutiny of their evaluation and audit practices.
  • ·The inclusion of compute governance and export controls as core governance topics underscores that trade compliance functions are now organizationally implicated in AI risk management, expanding the internal stakeholder set that must coordinate on AI governance strategy.

Stay ahead of stories like this

Get every ISO/OECD/UN AI governance development like this one, plus the rest of the week's developments. Every Thursday.

Free every Thursday. Unsubscribe anytime.

Governance controls affected

CHM-002Model Deployment Gate ProcessCMP-002International AI Standards Monitoring WorkflowCMP-005Regulatory Engagement Process for AI Standards DevelopmentHOC-001AI System Risk ClassificationMON-005Continuous Model EvaluationPRC-001AI Vendor Due DiligenceSAF-005Harmful Content Filtering

What to do now

  • ☐Review internal AI risk classification processes against the licensing and certification frameworks surveyed in the literature review to identify gaps ahead of potential pre-market approval obligations.
  • ☐Assign trade compliance teams to assess the organization's exposure to current and anticipated compute-related export controls, particularly those tightened by the United States and mirrored by allied governments.
  • ☐Evaluate existing system evaluation and conformity assessment procedures against the structured evaluation standards discussed in the review and referenced in the EU AI Act and NIST AI RMF.
  • ☐Brief procurement and legal counsel on the review's findings regarding AI procurement rules so that vendor contracts can be updated to reflect emerging government expectations around safety and transparency.
  • ☐Identify open research and regulatory questions highlighted in the review that are relevant to the organization's AI portfolio and determine where engagement with regulators or standards bodies would be productive.

What to watch next

Compliance teams should monitor legislative and regulatory developments in the United States, the United Kingdom, and the European Union related to compute export controls, as restrictions have been tightening rapidly and further allied government coordination is expected. Progress on formal AI certification schemes at standards bodies such as ISO and national equivalents warrants close attention, given that the review identifies certification as an area of active policy development that could result in binding pre-market obligations. Teams should also track any follow-on publications from LawAI and similar academic-policy institutions, as subsequent work addressing the open research gaps identified in this review may inform upcoming regulatory guidance cycles.

Related Coverage

Standards2026-11-04

2026 Midterm Shift Makes Federal Frontier AI Law Likely by Early 2027

Democrats swept the 2026 US midterm elections. Analysts at Transformer now assess that strong federal frontier AI legislation is significantly more likely to pass in early 2027. A bill exceeding the FRONTIER Act in stringency is anticipated, potentially including mandatory pre-deployment model testing and approval, strict liability for developers, and data center restrictions. Enterprise compliance teams that have treated federal AI regulation as a distant scenario must now treat it as a near-term planning constraint.

Corporate Policy2026-11-04

Google's AI Bundling Threat to Schools Exposes Coercive Adoption Risk

Google told K-12 schools they would lose Google Search access unless they opted in to AI features. This bundled essential service access with Gemini adoption. The plan drew antitrust criticism from lawmakers and former Federal Trade Commission Chair Lina Khan, who cited existing court findings of Google's monopoly liability. Following swift backlash, Google revised the plan so that Gemini, Gemini Notebook, and Search would each be off by default and independently switchable.

Research2026-11-03

AI-to-AI Weight Editing Breaks the 'Tested Model' Governance Assumption

MIT CSAIL researchers have demonstrated that GPT-6 Astra can directly rewrite the internal parameters of other AI models, boosting benchmark performance without conventional retraining. The finding undermines the core governance assumption that a model's behavior is fixed and verifiable after pre-deployment review. Organizations relying on one-time conformity assessments or static model audits face a structural gap in their AI assurance programs.