For the last two years, the AI story in finance has been one of acceleration: rapid movement from experimentation to pilots to early rollouts. APQC's newest research on AI in finance suggests that story is entering a new chapter. Adoption rates have leveled off year over year, and the conversation has shifted from whether to adopt AI to how to get real value out of it. For a benchmarking-minded reader, this is the more interesting question, because it's where measurable process improvement actually lives.
AI in Finance Adoption: A Plateau, Not a Ceiling
Between 2024 and 2025, finance functions moved quickly, embracing AI and generative AI to chase efficiency, accuracy, and better decision-making. By 2026, that growth curve has flattened. AI is now a baseline expectation rather than a novelty, and finance leaders have shifted their energy toward evaluating where it adds real value while keeping control, compliance, and risk management intact.
It would be a mistake, though, to read this plateau as a finish line. Many organizations are still in early maturity, deployments remain fragmented, and talent, integration, and governance challenges are unresolved. As generative and agentic AI capabilities mature, finance functions may well see a second wave of acceleration driven by deeper workflow integration and more autonomous decision support.
AI in Finance Today: Broad Deployment, Early Maturity
- Just over half of organizations have implemented or rolled out AI capabilities.
- Generative AI adoption continues to expand but is still in early maturity.
- Agentic AI is emerging — roughly half of organizations are exploring or implementing it.
The honest characterization of where finance sits today is scaled experimentation with targeted deployment, not full enterprise optimization. That distinction matters for anyone building a business case internally: broad deployment numbers can overstate how embedded, and how governed, these tools actually are.
AI in Finance Use Cases: Familiar Processes, Better Execution
One of the more telling findings is what hasn't changed. The top AI use cases in finance were the same in 2025 and 2026:
- Financial forecasting — steady at roughly half of organizations using AI here.
- Financial report generation — the second most common use case, though usage dipped from 48% to 42%.
- Market analysis — usage inched up from 39% to 41%.
- Quantitative analysis — holding fairly steady around 36–37%.
Rather than chasing new domains, organizations are doubling down on the processes where AI already delivers: high-volume, repeatable work with measurable cycle times — exactly the kind of process characteristics that make a business case easy to prove and easy to benchmark.
Generative AI in Finance Is the Workhorse; Agentic AI Is Still Being Tested
Generative AI has driven most of the adoption story so far, particularly for report generation and narrative creation. Agentic AI, systems capable of more autonomous execution, is now entering the picture, with about half of organizations reporting some level of implementation or rollout. But most of that activity remains exploratory. The shift toward more autonomous, real-time decision support in finance is directionally clear, even if it hasn't fully arrived yet.
Why AI in Finance Scaling Stalls: Talent, Change, and Process Gaps
The most significant barriers to scaling AI in finance are strikingly consistent year over year, and none of them are really about the technology itself:
- Lack of skilled talent — around 60% of organizations cite this, the single largest barrier.
- Employee resistance to change — in the mid-40% range.
- Challenges managing the transition to digital solutions — a persistent, if smaller, obstacle.
A talent gap is often a symptom rather than the root cause, it can trace back to process maturity, poor integration, or how work is designed in the first place. Scaling AI successfully depends less on better models and more on upskilling and reskilling, disciplined change management, and genuine alignment between technology, process, and people. Organizations that skip this groundwork risk stalling their AI initiatives no matter how much they invest in the tools themselves.
What AI in Finance Means for Finance Leaders
The arc from 2024 to 2026 is a shift from experimentation to operational integration. AI is no longer emerging in finance, it's foundational. But foundational doesn't mean mature. Most organizations are still working to scale and refine their use of AI across a narrow set of high-value processes, and the next phase will likely be defined by deeper integration into financial systems, greater reliance on generative and agentic capabilities, and a sharper focus on talent and organizational readiness.
There's also a compliance dimension worth watching. As AI becomes more embedded in financial reporting and decision-making, expect increasing regulatory, audit, and compliance scrutiny around transparency, controls, and accountability. A practical starting point for finance leaders: audit current AI skill requirements against existing workforce capabilities before expanding the scope of any AI initiative.
The Finance Benchmarking Opportunity: Measuring AI Maturity Against Peers
This is where the data gets useful beyond a single organization's dashboard. Adoption rate, maturity stage, use-case concentration, and the specific barriers slowing progress are all measurable, comparable data points, the raw material of good benchmarking. As finance functions move from proving that AI works to proving how well it works relative to peers, questions worth tracking include: How does our use-case concentration compare to the broader finance function? Where do we sit on the maturity curve relative to organizations of similar size and industry? And are our talent and change-management investments keeping pace with our deployment ambitions?
The organizations that treat this as a benchmarking question, not just a technology question, will be the ones positioned to move from scaled experimentation to genuine optimization.