Identifying hidden market exposure
The Pensions Regulator (TPR) has published guidance instructing pension scheme trustees to identify and manage hidden investment exposure to artificial intelligence (AI). As AI technology spreads across multiple industries, TPR warned that even apparently diversified portfolios could contain unaccounted concentration risks. TPR says trustees should examine their whole portfolio rather than evaluating individual funds or asset classes in isolation.
Rapid expansion of AI funding
The regulator noted that investment in AI-related activities is expanding rapidly across global markets. Reports cited by TPR, including the Stanford HAI 2026 AI Index Report, show that global private AI investment rose by 127.5% year-on-year in 2025, with total global corporate AI investment more than doubling over the same period. TPR highlighted that while this expansion presents new opportunities, it also introduces specific risks for pension scheme investments.
Five key steps for scheme governance
TPR expects schemes to maintain high standards of governance across short, medium, and long-term horizons. To meet these expectations, TPR expects trustees to assess direct and indirect AI exposure across all asset classes and perform scenario testing under severe but plausible stressed market conditions. Trustees must also review whether portfolio diversification remains effective if market correlations change during a downturn, challenge investment advisers on potential market impacts, and take proportionate action based on their scheme's individual circumstances.
Regulatory call to action
Ben Gunnee, TPR's Executive Director of Market Oversight, stated that trustees need to assess their exposure, test portfolio resilience, and question their advisers now. He stressed that schemes must understand how market developments could affect their investments under different market conditions.