Resources
Investor AI Hub curates resources on AI governance relevant to investors. Explore our guides to selected topics or browse the full resource library.
Group
All resources
78 results · Newest first
Company analysis › Frontier developers
AI control assessment
Guidelight produces standards covering control, capability testing, and transparency (Control, capability and transparency standards). Their first published assessment covered control practices (logging, monitor efficacy, gated actions, circuit breaking, third-party review, containment plans). No company scored above 3 of 5 on any practice and the grades ran from C+ (Anthropic, OpenAI) to F (Meta).
AI capabilities and scenarios › Single and catastrophic scenarios
AI 2040: Plan A
A detailed policy scenario created by the team behind AI 2027 exploring actions to delay the development of superintelligence until 2040.
Company analysis › Frontier developers
AI Safety Index - Summer 2026 Edition
A panel of seven independent experts grades nine companies across 37 indicators in six domains (risk assessment, current harms, safety frameworks, existential safety, governance and accountability, information sharing). Summer 2026 grades ranged from C+ (Anthropic) to F (xAI, DeepSeek, Mistral) and the index includes company-specific recommendations.
AI capabilities and scenarios › AI capability reviews and trends
Preliminary Report of the Independent International Scientific Panel on AI: Evidence-based assessment of opportunities, risks and impacts
A UN report by a panel of experts that covers AI capabilities, risks, access, market structure and governance. It is equivalent to the International AI Safety Report.
Company analysis › Frontier developers
SaferAI Ratings 2026
SaferAI scores twelve developers’ published frameworks on 65 weighted criteria across four dimensions: risk identification, risk analysis and evaluation, risk treatment, and risk governance. July 2026 scores ran from 35% (Anthropic) to 8% (Cohere). The methodology is published in Evaluating AI Providers’ Frontier Safety Frameworks and there is a companion report on emerging best practices (July 2026).
Investor frameworks and guidance › Investor publications
Global AI Governance: Frameworks in Formation
Report on evolving AI governance frameworks
AI capabilities and scenarios › Mainstream scenarios
AI Scenarios 2030: Helping policymakers plan for the future of AI
A set of five scenarios published by the UK government. The scenarios vary on six dimensions: capability, access, security, adoption, labour displacement and global cooperation. The scenarios are slow burn, open frontier, augmented growth, transformation economy, and take-off.
Investor frameworks and guidance › Investor publications
ESG Thema #23 - Digital Transition Series - The Social Dimensions of Responsible AI Deployment
A framework for investor stewardship on the social dimensions of AI
Company analysis › Frontier developers
xAI: The Unpriced Risk in SpaceX’s IPO
A group of AI governance non-profits published a report on xAI around the SpaceX IPO (Guidelight AI Standards, LASST, Encode AI, The Midas Project). This report included an assessment of four frontier AI developers against six critical AI governance practices.
AI capabilities and scenarios › Single and catastrophic scenarios
Europe 2031
A scenario published by the Arq Foundation on the risks to Europe from failing to catch up on AI.
Corporate AI governance › Corporate frameworks / guidance
Model AI Governance Framework for Agentic AI
Guidance for organisations on deploying agentic AI.
Corporate AI governance › Corporate frameworks / guidance
Corporate AI Risk Assessment Framework
The Corporate AI Risk Assessment Framework provides a basis for companies and external parties, including investors, to evaluate the management of AI-related risk.
Company analysis › Frontier developers
Frontier Risk Report
METR published a summary of its assessments of Anthropic, Google, Meta and OpenAI. Unlike the framework comparison above, this uses model access, questionnaires, agentic evaluations and red-teaming, with company review before publication. However, the report does not include the detailed company assessments.
Investor frameworks and guidance › Reviews of investor approaches
The Role of Investors in AI Governance
Review of the current approach of investors to AI governance and what would help.
AI capabilities and scenarios › AI capability reviews and trends
2026 AI Index Report
This is an annual report reviewing the state of the AI industry covering model performance; R&D; compute; economic impact; use in science, medicine, and education; and policy and governance. At 400+ pages it is a detailed review for investors
Investor frameworks and guidance › General
Briefing Paper: AI in Your Portfolio
Investor briefing on responsible AI
Investor frameworks and guidance › General
Responsible AI in practice: 2025 guidance for investors
Questions investors should ask on responsible AI
Investor frameworks and guidance › Investor publications
Responsible AI: Building trust and long-term value in the age of algorithms
Responsible AI investing framework
AI capabilities and scenarios › Reviews of scenarios and forecasts
Artificial General Intelligence Forecasting and Scenario Analysis
This report reviews AI forecasting and usefulness to policymakers. Chapter 4 - Understanding Disagreement: The Cruxes - is the most useful to investors as it outlines the three key questions that differentiate scenarios - capability, diffusion, and take-off.
Corporate AI governance › Disclosure frameworks and data
Governance that Performs: Oversight Risk and Return Signals in Tech
Analysis of AI governance of 52 companies
Corporate AI governance › Disclosure frameworks and data
Responsible AI in practice: 2025 global insights from the AI Company Data Initiative
Corporate survey based on UNESCO's Recommendation on the Ethics of AI.
Investor frameworks and guidance › Private markets
Responsible AI: The VC Perspective
Review of VC approach to responsible AI
Investor frameworks and guidance › Investor publications
AI Governance: Protecting portfolios in an AI driven world
Framework for AI governance
Corporate AI governance › Corporate frameworks / guidance
OECD Due Diligence Guidance for Responsible AI
This report provides practical guidance to enterprises for implementing OECD standards on responsible business conduct (RBC) and the OECD AI Principles when developing and using AI.
Investor frameworks and guidance › General
Investor Advocacy on Artificial Intelligence: A new Resource from SHARE
Investor briefing on responsible AI
Investor frameworks and guidance › Reviews of investor approaches
AI in the Boardroom 2025: Stewarding Workers’ Capital in the Age of System-Level AI Risk
Analysis of proxy voting on responsible AI resolutions in 2025.
Investor frameworks and guidance › Reviews of investor approaches
AI Stewardship in Motion: How First-Mover Asset Owners and Asset Managers are Responding to AI Risks
Review of investor approach to responsible AI in proxy voting and engagement.
AI capabilities and scenarios › Mainstream scenarios
Exploring possible AI trajectories through 2030
A set of four scenarios on the capabilities of AI published by the OECD. This has scenarios ranging from progress stalling to AI equalling or surpassing human capabilities across most or all activities by 2030. The report states all scenarios are plausible. The scenarios are focused on model capabilities, rather than other dimensions such as regulation, adoption and incidents.
AI capabilities and scenarios › AI capability reviews and trends
International AI Safety Report 2026
This is a regular review of AI capabilities and risks authored by over 100 AI experts. The 2026 version covers model capabilities, risks and governance.
AI capabilities and scenarios › Mainstream scenarios
Foresight on AI: Scenarios for an AI-enabled World
A report by the Government of Canada presenting four AI scenarios for policymakers. The scenarios focus on different paths of adoption and disruption in the economy, rather than model capabilities.
AI capabilities and scenarios › Mainstream scenarios
Global Economic and Financial Implications of Artificial Intelligence
A report by the IMF on two scenarios and the economic and financial implications - baseline and runaway diffusion. They cover the potential for strained government finances as labour taxes reduce and social spending increases.
AI capabilities and scenarios › Mainstream scenarios
The Futures of Artificial Intelligence
A report from the European Commission with five 2040 scenarios with an optimistic outlook for the EU. For example, one scenario is that the EU’s ‘safety first AI’ leads to a technological advantage.
AI capabilities and scenarios › AI capability reviews and trends
Frontier AI Trends Report
The UK AI Security Institute is a lead organisation conducting evaluations of models. This report presents trends in model capabilities and risks.
Investor frameworks and guidance › Reviews of investor approaches
Plugging the AI transparency gap: Collective Impact Coalition for Ethical AI 2025 Progress Report
Progress report for the WBA Collective Impact Coalition for Ethical Artificial Intelligence (AI CIC).
Company analysis › Frontier developers
Common Elements of Frontier AI Safety Policies
Periodically updated comparison by METR of 12 developers’ frameworks. A qualitative rather than quantitative assessment to identify the recurring elements. Currently includes capability thresholds, evaluations and elicitation, weight security, deployment mitigations, conditions for halting, evaluation timing, and oversight by boards or third parties.
Company analysis › Frontier developers
Foundation Model Transparency Index
This annual index scores 13 developers on 100 disclosure indicators across upstream, model and downstream domains. The 2025 average fell to 41 of 100 from 58 in 2024 (IBM 95; xAI and Midjourney 14).
Investor frameworks and guidance › Private markets
Responsible AI Due Diligence Tool
Due diligence tool for VC for responsible AI
AI capabilities and scenarios › AI capability reviews and trends
Second Key Update: Technical Safeguards and Risk Management
Update on technical approaches to general-purpose AI risk management.
Company analysis › AI safety model evaluations
HELM Safety Leaderboard
The Stanford Center for Research on Foundation Models publishes a model safety leaderboard based on five safety benchmarks. Stanford also publishes a benchmark on emerging government regulations (AIR-Bench, Nov 2025).
AI capabilities and scenarios › Reviews of scenarios and forecasts
The Longitudinal Expert AI Panel: Understanding Expert Views on AI Capabilities, Adoption, and Impact
A research project analysing expert predictions of AI. This provides investors with analysis on consensus expectations.
AI capabilities and scenarios › AI capability reviews and trends
First Key Update: Capabilities and Risk Implications
Update on capabilities and risks
AI capabilities and scenarios › Mainstream scenarios
Advanced AI: Possible futures: Five scenarios for how the AI-transition could unfold
A report with five AI scenarios from the Centre for Future Generations. The scenarios are plateau, big AI, diplomacy, arms race and take-off.
Investor frameworks and guidance › Investor publications
Achieving Effective AI Governance - From Theory to Practice
Framework for AI governance
Corporate AI governance › Corporate frameworks / guidance
EU AI Code of Practice
Code of practice for AI model developers
Investor frameworks and guidance › Investor publications
Governing the Algorithm: Investor Priorities for Responsible AI
Principles for approach to responsible AI
Company analysis › Digital rights and data privacy
RDR Index: Big Tech Edition
Ratings of 14 of the world's largest digital platforms.
AI capabilities and scenarios › Single and catastrophic scenarios
AI as Normal Technology
An essay arguing that AI should be thought of as a normal general-purpose technology like the internet or electricity.
AI capabilities and scenarios › Single and catastrophic scenarios
AI 2027
Presents a timeline from mid-2025 to the end of 2027 with a possible loss of control to AI. While many investors may find the ending extreme it outlines key tipping points and current developments are following its predictions.
AI capabilities and scenarios › Mainstream scenarios
How Artificial General Intelligence Could Affect the Rise and Fall of Nations
A report by RAND with eight scenarios on the geopolitics of AI.
AI capabilities and scenarios › AI capability reviews and trends
International AI Safety Report 2025
Comprehensive review of scientific research on the capabilities and risks of general-purpose AI systems.
AI capabilities and scenarios › Mainstream scenarios
Futures of Global AI Governance:
Four scenarios 1) Ideal scenario: Inclusive governance, share prosperity, 2) Suboptimal scenario: Cautious and controlled, 3) Suboptimal scenario: Fast and furious, and 4) Worst-case scenario: Breakthroughs backfire.
Investor frameworks and guidance › General
Responsible AI ESG Framework for investors
Responsible AI ESG framework for investors
Investor frameworks and guidance › Private markets
Human Rights Toolkit for Venture Capital General Partners
This resource from the UN B-Tech Project establishes a six-step framework designed to help venture capital (VC) general partners (GPs) understand how to assess technology startups’ human rights risks before investment and how to use their leverage to help portfolio startups identify and address these risks on an ongoing basis.
Investor frameworks and guidance › Private markets
Human Rights Toolkit for Venture Capital Limited Partners
This resource from the UN B-Tech Project establishes a six-step framework designed to help venture capital (VC) limited partners (LPs) understand how to assess general partners’ (GPs) approaches to technology startups’ human rights risks and how to encourage GPs to improve these approaches.
Investor frameworks and guidance › General
Responsible AI Playbook for Investors
Guide for investors on responsible AI engagement
Corporate AI governance › Principles
Principles for trustworthy AI
10 AI principles initially published in 2019 and updated in 2024.
Investor frameworks and guidance › General
Artificial Intelligence and Human Rights Investor Toolkit
RIAA guide for investors on financial and human rights risks from AI
Investor frameworks and guidance › General
Responsible AI Quick Guide for Asset Owners
Guide for asset owners on responsible AI
Investor frameworks and guidance › General
Artificial intelligence: An engagement guide
ICGN investor guide on AI
Investor frameworks and guidance › Private markets
Human Rights Risks in Tech: Engaging and Assessing Human Rights Risks Arising from Technology Company Business Models
This tool provides a series of “engagement cards” for investors to use in modeling their engagement with technology companies. Each card is a self-contained guide to engaging companies on a discrete issue connected to business model-related human rights risks.
Corporate AI governance › Corporate frameworks / guidance
ISO/IEC 42001:2023
International standard setting out requirements for establishing, implementing, maintaining and continually improving an AI management system.
Investor frameworks and guidance › Private markets
Responsible AI Startups (RAIS) Framework
Open-source resource aimed at fostering the safe and responsible application of AI. Based on the same resources Radical uses as part of its internal diligence and deliberation processes.
Corporate AI governance › Principles
Code of Conduct for Organizations Developing Advanced AI Systems
Code of conduct for advanced AI systems.
Company analysis › Digital rights and data privacy
WBA Digital Inclusion Benchmark
Assessment of 200 companies on digital inclusion using 15 indicators
Corporate AI governance › Corporate frameworks / guidance
NIST AI RMF 1.0
Voluntary US framework that helps organisations identify, assess and manage AI risks, organised around four functions: Govern, Map, Measure and Manage.
Corporate AI governance › Principles
Recommendation on the Ethics of Artificial Intelligence
Values, principles and policy recommendations.
Company analysis › Frontier developers
AI Policy Watchtower
Logs changes to, and breaches of, the safety policies, model cards and model specifications of OpenAI, Anthropic, Google, xAI, Meta, Microsoft and three smaller developers. It also tracks whether each change was disclosed. This grew out of the project’s Seoul Commitment Tracker (February 2025), which found six of the original sixteen signatories had not met their Seoul commitments, although this has now been surpassed by legislation.
Company analysis › AI safety model evaluations
AI Risk Assessments
Common Sense Media publishes risk assessments of AI models for use by children
Company analysis › AI safety model evaluations
AI Security Leaderboard
FAR.AI has a leaderboard assessing how hard it is to make frontier AI models produce harmful content (jailbreak).
Corporate AI governance › Corporate frameworks / guidance
AI Verify Testing Framework
Helps companies assess the responsible implementation of their AI system against 11 internationally recognised AI governance principles.
Company analysis › AI safety model evaluations
CAIS AI Dashboard
This is a dashboard of model evaluations covering a range of areas including a Risk index section.
AI capabilities and scenarios › Reviews of scenarios and forecasts
Forecasting platform covering AI
A platform that aggregates forecasts from individuals covering a range of AI questions. Questions include a probability and comments with useful references.
Corporate AI governance › Disclosure frameworks and data
Hiroshima AI Process Reporting Framework
The Hiroshima AI Process Reporting Framework, run by the OECD since 2025, is a voluntary transparency mechanism for organisations developing advanced AI systems to publicly report how they follow the G7 Hiroshima Process International Code of Conduct.
Explore the full library
Sign in or apply for membership to explore member resources.