Workiva study highlights AI governance risks in France

As artificial intelligence moves from assisting employees to executing autonomous workflows, the need for robust AI governance is increasing, particularly in regulated sectors like finance and audit. Experts emphasize that for AI agents to be integrated into reporting and compliance, companies must be able to trace information sources, identify approvals, and reconstruct actions taken by an agent on their behalf.

A Workiva study involving French executives highlights the practical risks of this transition. According to the 2026 Midyear Executive Benchmark, 26% of French leaders reported that internal AI audits detected errors that reached external audiences or boards of directors. Furthermore, 48% of French respondents expressed only moderate confidence in AI outputs, citing a need for human supervision and data traceability.

Issues with data fragmentation and inconsistent definitions remain significant hurdles. In critical areas such as financial reporting and sustainability disclosures, AI errors can lead to regulatory, reputational, and legal consequences. Experts suggest that relying on experienced professionals to spot anomalies is insufficient; instead, organizations require systematic governance models to ensure AI actions are justifiable and verifiable.