> **来源：[研报客](https://pc.yanbaoke.cn)** # KPMG Global AI in Finance 2026 Summary ## Core Content KPMG's *AI in Finance 2026* report explores how AI is transforming the finance function, emphasizing its role as a **decision-engine** rather than a **cost lever**. The findings are based on a survey of 1,013 senior finance leaders across 13 sectors and 20 countries, highlighting the **maturity gap** between AI adoption and performance realization. --- ## Main Findings - **AI Adoption Growth**: Active AI use across the finance function has more than doubled in two years, with over 75% of organizations now using AI in financial planning, reporting, and commercial analysis. - **Performance Gains**: The strongest performance improvements are in **judgment-heavy areas** such as decision-making quality, forecast accuracy, and responsiveness. Agentic AI, which can plan, reason, and act autonomously, delivers the most significant gains, with organizations using it reporting **32% to 40%** stronger performance in these areas. - **Sector Differences**: There are notable gaps in performance between sectors. **Banking** and **Technology** lead in areas like close efficiency and forecast accuracy, while **Healthcare** and **Consumer** lag significantly, particularly in forecast accuracy (27-point gap) and ROI. - **ROI Expectations**: 71% of organizations report that AI is meeting or exceeding ROI expectations, but only **23%** report it exceeding expectations in key metrics, indicating a disparity between adoption and impact. --- ## Key Themes ### 1. **AI as a Decision-Engine** - AI is not just about automation or cost reduction. It's enhancing **judgment and decision-making** in finance. - Organizations that deploy agentic AI see **stronger performance gains**, especially in areas that require strategic thinking and risk assessment. - The real value of AI lies in **sharpening human judgment**, not merely speeding up tasks. ### 2. **Governance and Controls Build Confidence** - Strong governance, controls, and human oversight are critical for **trust and scalability** in AI deployment. - **Assurance-ready** organizations report **3 to 6 times** the rate of significant improvement in key metrics. - The ability to produce **audit trails and evidence** is a key differentiator for high-performing finance functions. ### 3. **The Assurance Readiness Gap** - Only **42%** of all organizations are strongly assurance-ready for AI-enabled finance processes. - **60%** of agentic AI leaders are assurance-ready, but a significant gap remains. - Assurance readiness is no longer an internal concern but a **commercial and regulatory requirement**. ### 4. **Data Quality and Workforce Gap** - **Data quality** is the most cited barrier and opportunity for AI in finance. - Organizations are struggling with **fragmented data sources**, **slow integrations**, and **legacy systems**. - Only **28%** of organizations are rethinking the **types of talent** they need, indicating a **workforce gap** in AI capabilities. - **Human oversight** remains crucial, especially in ensuring **trust in AI outputs**. --- ## Sector Performance Comparison | Metric | Lowest Performing Sector | Highest Performing Sector | |----------------------|--------------------------|---------------------------| | Close Efficiency | Healthcare (47%) | Banking (76%) | | Forecast Accuracy | Healthcare (44%) | Banking (71%) | | ROI | Healthcare (47%) | Banking (70%) | | Decision Quality | Healthcare (62%) | Banking (71%) | | Decision Speed | Healthcare (50%) | Banking (65%) | | Error Reduction | Healthcare (50%) | Banking (65%) | --- ## Recommendations for Finance Leaders - **Focus on Judgment-Heavy Work**: Prioritize areas where AI can enhance decision-making, not just automate routine tasks. - **Invest in Governance and Controls**: Build a **Trusted AI framework** that includes **fairness, transparency, explainability, accountability**, and **data integrity**. - **Track AI-Related KPIs**: Formal tracking of AI performance metrics is a strong indicator of success. - **Develop a Total Workforce Model**: Move beyond training to rethinking the **human-AI operating model**. - **Address Data Quality**: Focus on **cleaning priority data** for AI use cases, not the entire data estate. --- ## KPMG's Role KPMG is helping organizations: - Modernize their **data estates**. - Implement **governance and control frameworks**. - Build **assurance readiness** through **risk assessments** and **audit trails**. - Scale AI responsibly with **ethical AI principles** and **human oversight**. --- ## Conclusion AI in finance is moving beyond **efficiency** and into **decision-making**. The key to success lies in **trust**, **governance**, and **judgment enhancement**. While adoption is widespread, performance is uneven, with **Banking and Technology** leading the way. The **assurance readiness gap** and **data quality issues** remain critical challenges, especially for sectors like **Healthcare** and **Consumer**. KPMG emphasizes the need for a **holistic approach** to AI in finance, integrating **technology, people, and processes** to achieve sustainable value.