UK pension fund manager Railpen has warned that companies' exposure to AI-related risks is increasing faster than governance and risk management practices are evolving, creating what it describes as an "urgent governance challenge" for investors. A new report from the organisation also highlights the role investors can play in strengthening AI governance expectations and oversight.

The findings add to growing investor concern that corporate AI adoption is advancing more quickly than the governance frameworks needed to identify, manage and disclose associated risks. As AI becomes increasingly embedded across business operations, investors face mounting pressure to assess whether boards have adequate oversight of the technology's risks and opportunities.

Investor role in advancing AI governance

Railpen's research found that many companies' AI governance frameworks remain less developed than their exposure to AI-related risks, raising questions around risk management, board oversight and disclosure practices.

According to the report, investors have an important role to play in driving improvements. It highlights the use of structured assessments to identify AI-related risks and evaluate governance practices across portfolios, alongside engagement with companies to strengthen governance and disclosure standards. The report also points to the importance of sharing examples of good practice and promoting common expectations across the investment and policymaking landscape.

While leading approaches to AI governance and risk management are beginning to emerge, Railpen notes that many companies have yet to move beyond high-level commitments and demonstrate how governance frameworks are being implemented in practice. The report also finds that AI-related risks are often treated as an extension of existing risk categories, such as cybersecurity, rather than as a source of distinct organisational, operational and governance challenges.

"Managing the risks and the opportunities posed by AI requires a system-wide approach," said Caroline Escott, Head of Investment Stewardship and Co-Head of Sustainable Ownership at Railpen. "Investors should be looking to actively engage companies to make sure they are using best AI practices. By uniting around a common set of expectations, investors can work collectively to reinforce efforts to drive responsible AI use."

Growing AI governance concerns

Responsible AI governance is a priority for Minerva Analytics. Last year, Minerva rolled out additional research and voting guidelines to evaluate corporate disclosures against globally recognised cyber governance standards such as the OECD AI Principles and the G7 Hiroshima AI Process. These new guidelines supplemented Minerva’s existing cyber-governance questions first adopted in 2016, offering investors a robust lens through which to assess board readiness with a clear focus on governance and disclosure quality, particularly in key regulatory disclosures such as annual reports, CSR disclosures, as well as corporate websites.

AI governance is also receiving increasing attention from policymakers, regulators and international institutions. This week, UN High Commissioner for Human Rights Volker Türk called on states and leading frontier AI companies to take meaningful action to ensure governance frameworks keep pace with the technology's development and associated risks to human rights. UNESCO has also announced new tools to support ethical AI governance at the fourth UNESCO Global Forum on the Ethics of AI 2026.

The issue has attracted particular scrutiny in recent weeks following warnings from Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman and current and former employees of both companies regarding the speed of AI development and the potential consequences if governance and safeguards fail to keep pace. Growing attention on AI-related risks comes as the sector continues to expand rapidly, with Anthropic reportedly preparing for an IPO and OpenAI also widely expected to pursue a public listing.

Railpen's latest report builds on its 2025 publication Achieving Effective AI Governance, which introduced the AI Governance Framework (AIGF) designed to translate responsible AI principles into practical governance measures. Applying the framework to companies in Railpen's Active Equity portfolio, the latest research found that AI governance maturity remains uneven and frequently underdeveloped relative to risk exposure. The report notes that relatively few companies provide detailed disclosure on how AI risks are assessed, managed and overseen, suggesting investors are likely to continue encountering significant variation in governance quality even among companies with similar exposure to AI technologies.

Key takeaways

Railpen's findings reinforce the view that AI governance is moving from a specialist technology issue to a core investment and stewardship consideration. While AI adoption continues to accelerate, governance frameworks and disclosures often lag behind, leaving investors with limited visibility over how risks are identified, managed and overseen.

For investors, the report underlines the importance of scrutinising AI governance practices alongside broader risk management and board oversight arrangements. As AI-related opportunities and exposures continue to expand, expectations around governance quality, accountability and disclosure are likely to become increasingly prominent features of investment stewardship and corporate engagement.