In this episode, Cambria Allen-Ratzlaff, Interim CEO of the PRI, is joined by Michael Benedict Yamoah (Vice President, Stewardship Director, EOS at Federated Hermes), Chris Jurgens (Senior Director, Omidyar Network), and Oumou Ly (Non-resident Research Fellow, UC Berkeley Center for Long-Term Cybersecurity) to explore how investors should respond to AI.

Building on Part 1, this episode moves from theory to practice, outlining how investors can assess AI governance, identify risks across portfolios, and begin engaging with companies in a fast-moving and uncertain landscape.

Overview:

AI is already reshaping portfolios, but most investors are still early in understanding how to manage the risks. This episode focuses on practical steps, from governance and engagement to tools, research, frameworks and real-world examples of leading practice.

A key message is that there is no perfect framework yet. Instead, investors must start now, build capability over time, and engage continuously as the technology evolves.

Detailed coverage:

What good AI governance looks like
At a minimum, companies must comply with regulation and establish clear internal policies. Strong governance goes further, embedding AI into enterprise risk management, assigning board-level responsibility, and ensuring oversight across the organisation.

Beyond compliance: lifecycle thinking
Investors are encouraged to assess the full lifecycle of AI systems, from development and deployment to real-world impacts, liabilities and societal consequences.

AI risk is dynamic
Unlike other technologies, AI systems evolve post-deployment. This requires continuous monitoring, disclosure and adaptation, rather than one-off assessments.

Examples of leading practice
Companies such as Anthropic and Microsoft are highlighted for transparency, investor engagement and responsible AI frameworks. Across the ecosystem, progress is being driven by collaboration between companies, investors and policymakers.

The importance of infrastructure and ecosystems
AI is not just about software, it spans chips, data centres and energy systems. Managing its risks requires coordination across the full value chain.

Practical starting points for investors
Investors should map where AI sits in their portfolios, identify key use cases, and assess associated risks such as cybersecurity, compliance and liability.

Tools, frameworks and collaboration
A growing ecosystem of resources, from investor coalitions to research frameworks, is emerging to support engagement and analysis.

A marathon, not a sprint
AI governance is an ongoing process. Investors must build long-term capability, stay engaged in dialogue, and avoid waiting for perfect solutions before acting.

Start now, signal intent
Even simple engagement, asking basic governance questions, can send a strong signal to companies that responsible AI matters.

Chapters:

00:08 - Introduction: from AI risk to investor action
01:00 - What good AI governance looks like
03:05 - Internal policies, risk management and board oversight
05:00 - Lifecycle thinking and real-world impacts
08:17 - Examples of leading practice in AI governance
10:30 - Defining and understanding AI risk
13:15 - Mapping AI use cases across portfolios
15:39 - Practical tools and investor resources
19:44 - Why AI is a marathon, not a sprint
22:24 - Final takeaways: start now and engage

Further reading: Anthropic labor market impacts, Microsoft transparency report

Disclaimer:

This podcast and material referenced herein is provided for information only. It is not intended to be investment, legal, tax or other advice, nor is it intended to be relied upon in making an investment or other decision. PRI Association is not responsible for any decision made or action taken based on information on this podcast. Listeners retain sole discretion over whether and how to use the information contained herein. PRI Association is not responsible for and does not endorse third parties featured on in this podcast or any third-party comments, content or other resources that may be included or referenced herein. Unless otherwise stated, podcast content does not necessarily represent the views of signatories to the Principles for Responsible Investment. All information is provided “as is” with no guarantee of completeness, accuracy or timeliness, or of the results obtained from the use of this information, and without warranty of any kind, expressed or implied. PRI Association is committed to compliance with all applicable laws. Copyright © PRI Association 2026. All rights reserved. This content may not be reproduced, or used for any other purpose, without the prior written consent of PRI Association.