AI cannot replace XBRL—but it can reduce the tagging burden and improve reporting quality

The We Mean Business Coalition advocates for the retention of XBRL (eXtensible Business Reporting Language) tagging for financial and non-financial reports, opposing proposals by the European Commission to simplify regulation by removing these requirements. The authors argue that while AI can automate tagging, detect errors, and improve data extraction, it cannot replace XBRL because AI operates probabilistically and lacks the standardization, regulatory compliance, and data integrity required for reliable analysis. The document emphasizes that relying solely on AI-extracted data from PDFs risks inaccuracies in KPIs and figures, which could distort investor analyses and market trust. It concludes that AI should be leveraged to modernize and accelerate the tagging process, but the rule-based structure of XBRL and the associated assurance process by auditors remain essential for traceable and explainable reporting.

• AI is incapable of replacing XBRL because it lacks the necessary standardization, regulatory compliance, and structured data integrity required for reporting.
• AI can optimize XBRL reporting by automating the mapping of data to taxonomies, detecting anomalies and errors before submission, extracting data from various formats, and enhancing analytics for stakeholders.
• Relying exclusively on AI to extract data from PDF reports introduces significant risks of inaccuracies, such as incorrect units or misassigned KPIs, which could negatively impact stock prices and market trust.
• AI cannot substitute for the assurance process provided by auditors and senior management, nor can it fully handle complex reporting needs or new regulations due to its reliance on past data.
• The 'black box' nature of complex AI algorithms conflicts with the demands of regulators and investors for reporting that is traceable and explainable.

APA

Jagd, J. T., & Fischbach, S. (2025). AI cannot replace XBRL—but it can reduce the tagging burden and improve reporting quality. We Mean Business Coalition. https://www.wemeanbusinesscoalition.org/blog/ai-cannot-replace-xbrl-but-it-can-reduce-the-tagging-burden-and-improve-reporting-quality/

Chicago

Jagd, Jane Thostrup, and Sebastian Fischbach. AI cannot replace XBRL—but it can reduce the tagging burden and improve reporting quality. We Mean Business Coalition, 2025. https://www.wemeanbusinesscoalition.org/blog/ai-cannot-replace-xbrl-but-it-can-reduce-the-tagging-burden-and-improve-reporting-quality/.

Wikipedia

{{cite press release |last1=Jagd |first1=Jane Thostrup |last2=Fischbach |first2=Sebastian |title=AI cannot replace XBRL—but it can reduce the tagging burden and improve reporting quality |publisher=We Mean Business Coalition |date=10 February 2025 |url=https://www.wemeanbusinesscoalition.org/blog/ai-cannot-replace-xbrl-but-it-can-reduce-the-tagging-burden-and-improve-reporting-quality/ |access-date=23 October 2026 |via=Climate Insights Directory}}

BibTeX

@misc{jagd2025cannot, author = {Jagd, Jane Thostrup and Fischbach, Sebastian}, title = {{AI cannot replace XBRL—but it can reduce the tagging burden and improve reporting quality}}, publisher = {We Mean Business Coalition}, year = {2025}, month = feb, url = {https://www.wemeanbusinesscoalition.org/blog/ai-cannot-replace-xbrl-but-it-can-reduce-the-tagging-burden-and-improve-reporting-quality/}, urldate = {2026-10-23}, note = {Indexed by Climate Insights Directory} }