Global Study Reveals Biggest Risks of AI in Finance Sector

Why It Matters

The findings highlight a widening gap between rapid AI adoption in finance and lagging regulatory oversight, exposing the sector to privacy breaches, model failures, and systemic risk. Addressing these gaps is crucial to safeguard consumer protection and maintain market stability as AI becomes integral to financial operations.

Key Takeaways

• Data privacy cited as top AI risk by 80% of regulators
• AI hallucinations concern 70% of surveyed financial authorities
• 80% of firms have deployed AI, but 48% of regulators still exploring
• 42% see AI fighting financial crime; 18% think it worsens it
• Governance gaps persist as UK MPs label regulator approach “wait‑and‑see”

Pulse Analysis

The Cambridge Judge Business School’s global survey of 628 finance organisations paints a vivid picture of AI’s double‑edged sword in the sector. While 80% of firms report some level of AI deployment—particularly in software development—regulators are grappling with the technology’s darker side. Data privacy emerges as the most cited threat, reflecting heightened sensitivity to cross‑border data flows and customer confidentiality. Equally concerning are AI hallucinations, where models generate inaccurate or fabricated outputs, potentially distorting risk assessments, credit decisions, and market forecasts.

Regulatory readiness lags behind adoption, with 48% of the 130 surveyed authorities still in an exploratory phase. In the UK, the Treasury Committee’s recent report slammed the Bank of England and the FCA for a “wait‑and‑see” stance, warning that insufficient oversight could erode consumer trust and destabilize the financial system. Similar sentiments echo across the globe, as central banks and international bodies like the BIS and IMF join the dialogue. The study underscores a fragmented governance landscape, where accountability for AI‑driven errors remains unclear, and cyber‑related vulnerabilities outpace human supervision.

For industry leaders, the message is clear: robust AI governance is no longer optional. Firms must translate high‑level regulatory principles into day‑to‑day controls, invest in explainable AI, and embed continuous monitoring to mitigate bias and opacity. As 78% of regulators anticipate AI’s transformative impact by 2030, those that proactively address risk management will capture productivity gains while safeguarding financial inclusion and crime‑fighting capabilities. The coming years will likely see tighter standards, cross‑border coordination, and a shift from exploratory pilots to mature, compliant AI ecosystems.