Poor data quality has limited AI use in financial and sustainability reporting for 71% of surveyed executives, while 89% of institutional investors are concerned about AI accuracy in company disclosures.

Key takeaways

• 84% of surveyed executives said they were somewhat (45%) to very confident (39%) that AI-generated information could appear in an annual report without human review.
• 26% said internal audits detected AI errors that reached external audiences or the board.

• 71% of surveyed executives said poor data quality had limited AI use in financial and sustainability reporting, while 47% of institutional investors said they actively look for AI-generated errors in company disclosures.

Workiva’s 2026 Midyear Executive Benchmark Survey found a wide gap between executives’ confidence in unreviewed AI material and the AI errors their organizations identified.

The survey found that 84% of executives were at least somewhat confident in the accuracy of AI output appearing in an annual report without human review. At the same time, 26% said internal AI audits detected errors that reached external audiences or the board.

Executives trust AI output despite detected errors

The 84% confidence figure included 39% of executives who said they were very confident and 45% who were somewhat confident in unreviewed AI-generated information.

The survey also asked professionals below the C-suite the same question. Among that group, 76% were at least somewhat confident, including 29% who were very confident.

The report does not say that all detected AI errors appeared in annual reports. It also does not describe the type or severity of the errors that reached outside audiences or boards.

Poor data quality limits AI reporting

Only 11% of executives said their organizations’ data quality was sufficient for AI use.

Another 71% said poor data quality had at least moderately limited AI use in financial and sustainability reporting. That included 27% who said poor data had substantially blocked AI use in important business processes.

Investors scrutinize AI-generated disclosures

Investors are paying close attention. In the survey, 89% said they were concerned about AI errors in company disclosures, and 47% said they actively check those disclosures for signs of AI-generated mistakes.

The survey also found that 96% considered company policies for AI governance and human oversight important to their investment decisions. That included 62% who considered those policies very important.

Workiva surveyed 2,272 finance, risk, sustainability, and legal professionals, including 847 C-level executives, plus 367 institutional investors. The investor sample covered the United States, Canada, and the United Kingdom. The professional respondents worked at organizations with at least 250 employees or $250 million in annual revenue across North America, Latin America, Europe, and the Asia-Pacific region.