BDO Research: AI Moves Into High-Stakes Financial Judgments, Raising the Bar for Governance

Press Release alarm

Finance leaders are confident in AI's role in the audit, yet governance and documentation still lag behind adoption.

Report Highlights:

  • AI adoption is near universal. Nearly every finance leader surveyed (99%) reports that their organization uses AI to generate or inform at least one financial estimate or judgment reviewed by its external auditor. Applications include cash flow or earnings forecasts (62%), tax provisions or deferred tax calculations (55%), and fair value measurements or asset valuations (54%).
  • Agentic AI in finance is rising. Nearly half (47%) of respondents are already piloting or deploying agentic AI, and a similar share (46%) is actively evaluating it for the finance function.
  • Governance maturity is uneven. Less than half of respondents (49%) say they have the documentation and processes needed to explain and substantiate all AI-assisted outputs, while another 46% can explain most outputs but still face gaps. Closing that supportability gap will require strong data practices, clear ownership, reliable controls, consistent documentation, and effective governance.


CHICAGO, November 4, 2026 — BDO’s third annual Audit Innovation Survey, released today, finds that AI has moved beyond operational and analytical tasks and into the financial estimates and judgments that external auditors review. The research also finds that confidence in AI is outpacing the governance, documentation, and evidence needed to fully support AI-informed results.

Nearly every finance leader surveyed (99%) reports that their organization uses AI to generate or inform at least one financial estimate or judgment reviewed by its external auditor. The most common applications include cash flow or earnings forecasts (62%), tax provisions or deferred tax calculations (55%), and fair value measurements or asset valuations (54%). AI is also being applied to credit loss allowances (50%), liability estimates (47%), revenue recognition judgments (46%), and impairment assessments (42%).

The conversation has shifted from ‘What technology are you using?’ to ‘Can I trust the outcome?' Finance leaders increasingly expect modern audit tools and AI-enabled capabilities, but technology alone is not what builds confidence. The future of assurance is not defined by the technology we use, but by the confidence we create. Clients choose and retain firms based on trust in their people, the strength of their relationships, and the ability to deliver quality outcomes.

Demetrios Frangiskatos
BDO Assurance National Managing Principal

AI ambitions are high, but governance maturity is uneven

Although AI is widely used within the finance function, less than half of respondents (49%) say they have the documentation and processes needed to explain and substantiate all AI-assisted outputs, while another 46% can explain most outputs but still face gaps.

Data governance maturity varies. Nearly half of respondents (49%) describe their finance department's data governance as managed or optimized, and 29% have formal policies in place but still rely on some manual reconciliation. Roughly one in five characterize their practices as developing or ad hoc, and 17% say their AI ambitions are advancing faster than the data infrastructure and governance needed to support them.

Despite these governance gaps, more than half (55%) of respondents report a defined finance and accounting AI roadmap that includes both routine tasks and significant estimates, backed by formal investment commitments and milestones. Nearly half (47%) are already piloting or deploying agentic AI, and a similar share (46%) is actively evaluating it for the finance function.

The findings point to a gap between enthusiasm for AI and the readiness to support it. Closing the AI supportability gap means creating strong data practices, clear ownership, reliable controls, consistent documentation, and effective governance.


AI is reshaping both audit capabilities and risks

Nearly two-thirds of finance leaders (65%) expect their auditors’ use of AI to deliver more rigorous audits. They also see more time spent evaluating complex risks and unusual transactions as the top benefit from auditors’ use of AI. But when companies use AI in significant financial estimates without the ability to fully support those estimates, new risks can be introduced into the audit. Auditors may then need to spend more effort addressing AI-related risks, which limits how much of the time saved through AI they can dedicate to complex risks and unusual transactions.

Finance leaders’ expectation of more rigorous audits comes with a condition: Eighty-six percent worry that overreliance on AI could reduce human judgment in the audit. They want AI to strengthen auditor expertise, not replace it. Inaccurate outputs, cybersecurity, data privacy, ungoverned use, bias, and auditability also rank as prominent concerns.


Continuous auditing interest exceeds readiness

Finance leaders see value in more continuous, year-round auditing, citing potential benefits such as earlier risk identification, fewer year-end surprises, and more timely insight into issues affecting financial reporting. Yet only 6% say their organization is ready today. Respondents name better internal data infrastructure (71%), stronger AI governance (63%), clearer regulatory guidance (62%), greater auditor capability in continuous auditing methods (57%), and lower cost (48%) as prerequisites. The report suggests these foundations are also needed to support AI-informed estimates in the current model, making continuous audit readiness an extension of the work organizations already need to do to close the gap between AI adoption and AI governance. 

Download the full BDO Audit Innovation Survey here.

Media contact: 
Marisha Chinsky, Highwire 
Email