AI Governance Institute logo
AI Governance Institute

Practical Governance for Enterprise AI

← News
Research2026-04-19

Seven Major AI Companies Rated Across 33 Safety Indicators in Future of Life Institute's Summer 2025 Index

What happened

The Future of Life Institute published its Summer 2025 AI Safety Index on July 15, 2025, evaluating seven leading AI companies against 33 indicators of responsible AI development organized into six domains, including risk ownership, accountability, and oversight. The index is global in scope and does not publicly identify all evaluated companies in its raw findings, but it singles out DeepMind with specific recommendations covering better coordination between safety and policy teams, greater transparency in third-party evaluations, and publication of risk assessments in model cards. The report identifies persistent gaps between corporate safety commitments and actual operational practices, reflecting growing concern among regulators in the European Union, United Kingdom, and United States about whether voluntary frameworks correspond to measurable internal controls. The Future of Life Institute's methodology across 33 discrete indicators represents an industry-wide push to move AI governance assessment beyond high-level policy statements toward auditable, documented evidence of practice. For enterprise compliance teams, the index functions as an external benchmark that regulators, investors, and procurement officers may reference when assessing vendor AI governance maturity.

Why it matters

  • ·Regulators in the EU, UK, and US have signaled interest in whether voluntary AI safety frameworks translate into verifiable controls, meaning this index could be cited as evidence in supervisory reviews or enforcement proceedings, increasing regulatory exposure for companies that score poorly or fail to address identified gaps.
  • ·The specific recommendations directed at DeepMind around model cards and third-party audit transparency signal areas where documentation standards may tighten industry-wide, requiring procurement and compliance teams to revisit vendor contracts for provisions covering model documentation and audit rights.
  • ·Organizations procuring AI systems from any of the seven evaluated companies face reputational and third-party risk if vendor governance maturity is rated poorly, as investors and procurement officers may use the index as a qualification filter that indirectly affects supplier relationships and contract renewals.

Governance controls affected

CMP-008Federal AI Regulatory Monitoring and Pre-Deployment VettingHOC-001AI Risk ClassificationMON-005Model Card and Documentation MaintenancePRC-001Third-Party AI Risk AssessmentPRC-002AI Vendor Contract RequirementsPRC-004Vendor Incident Notification Requirements

What to do now

  • ☐Review existing vendor contracts with any of the seven evaluated AI companies to confirm provisions for model documentation, audit rights, and third-party evaluation transparency are present and enforceable.
  • ☐Cross-reference your organization's model card templates against the Summer 2025 AI Safety Index recommendations to identify documentation gaps, particularly around published risk assessments.
  • ☐Update your third-party AI risk assessment process to incorporate the index's 33 indicators as a structured checklist when onboarding or re-evaluating AI vendors.
  • ☐Brief procurement and legal teams on the index findings so that future AI supplier qualification processes can account for vendor safety ratings and any outstanding remediation commitments.
  • ☐Establish a monitoring cadence to track whether DeepMind and other evaluated companies publish updated model cards or third-party audit disclosures in response to the index recommendations, and record findings in your vendor risk register.

What to watch next

Compliance teams should monitor whether future editions of the Future of Life Institute's AI Safety Index, expected to continue on a recurring basis, begin establishing de facto disclosure expectations that regulators in the EU, UK, or US formally reference in guidance or enforcement actions. Teams should also track whether DeepMind or other evaluated companies issue updated model cards, revised third-party audit disclosures, or public responses to the index recommendations, as these outputs may signal shifting industry documentation norms. Any regulatory body that cites the index in rulemaking or supervisory communications would represent a significant signal that voluntary benchmarking is transitioning toward a compliance baseline.

Related Coverage

Insight2026-06-13

Fable 5 and Mythos 5 Suspended by U.S. Export Control Directive: Three Governance Gaps Enterprise AI Programs Have Not Planned For

On June 12, 2026, a U.S. government export control directive required Anthropic to suspend all access to Fable 5 and Mythos 5 for foreign nationals, effectively disabling the models for all customers overnight. The immediate trigger was a narrow code-analysis jailbreak technique, but the directive exposes deeper gaps: most enterprise AI governance programs have no continuity plan for government-mandated model suspension, no process for nationality-based access controls, and no export control review in their AI vendor assessment workflow.

Insight2026-06-27

Mythos 5 Partial Reinstatement Creates Government-Controlled AI Access Tiers With No Transparent Process

The US government on June 27 granted roughly 100 approved companies access to Claude Mythos 5, partially reversing a June 12 export control suspension, while Fable 5 and organizations outside the approved list remain locked out with no published selection criteria or recourse. The action is the first commercial enforcement under a new executive order framework requiring government pre-release review of frontier models, making tiered access structural rather than ad hoc.

Research2026-06-15

S&P Global Report Frames AI Governance as a Principle-Based Risk Discipline, Raising the Bar for Enterprise Compliance Programs

S&P Global has published a research report titled 'The AI Governance Challenge,' arguing that enterprise AI governance should be anchored in five core principles: transparency, fairness, privacy, adaptability, and accountability. The report documents common organizational practices including ethical review boards, impact assessments, algorithmic transparency mechanisms, and risk-focused controls. Its findings map directly to compliance, model governance, and privacy programs across industries.