The latest insights from the We Mean Business coalition

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

The European Commission’s so-called Omnibus proposals, a package of measures intended to simplify regulation, have recently caused widespread concern among the business community. While some simplification is necessary and helpful, the Commission should not throw the baby out with the bathwater by allowing deregulation and removal of essential parts of the legislation itself. For example, we have heard increasing calls to eliminate the requirement for XBRL (eXtensible Business Reporting Language) tagging of companies’ financial and non-financial reports and for assurance of these XBRL-based reports. These digital reports have been mandated in the EU – and other jurisdictions – as they make the financial and non-financial data available in a structured format, which is suitable for analyses.

The argument now being made is that artificial intelligence (AI) should enable investors and other capital providers to extract the digital data they need independently, some even claim more accurately and efficiently, allowing companies to return to publishing reports in simple PDF formats. 

But that is not entirely true. 

While AI is a transformative force in business and finance, it cannot replace XBRL because it lacks standardization, regulatory compliance, and structured data integrity. However, AI can significantly enhance how companies prepare, validate, and analyze XBRL reports, making the process faster, more accurate, and more insightful.

How AI can support XBRL reporting 

Rather than replacing XBRL, AI can be a powerful tool in optimizing and simplifying XBRL reporting by: 

• Automating data tagging – AI can classify and map financial and non-financial data to the correct XBRL taxonomy. 

• Detecting errors and inconsistencies – AI-driven anomaly detection can flag reporting mistakes before submission. 

• Improving data extraction – AI can extract financial and non-financial data from different formats and convert them into structured XBRL reports. 

• Enhancing analytics – AI can analyze XBRL reports, providing deeper insights for regulators, investors, and analysts. 

Why XBRL remains essential 

Regulatory bodies worldwide mandate XBRL-based reporting to ensure uniformity in financial and non-financial disclosures. This standardization ensures that reported data is useful for analysis by national business authorities, investors, and other stakeholders. It also ensures that all companies comply with reporting regulations consistently, making it easier to compare performance and prospects. 

AI, on the other hand, operates probabilistically—it learns patterns and makes predictions, but it does not guarantee compliance with strict financial and non-financial reporting rules. If national business authorities, investors and other stakeholders rely solely on AI-extracted data, there is a significant risk of inaccuracies—including misassigned KPIs, incorrect units, and erroneous figures. These errors can for instance distort investor analyses of a given company, with potential consequences for stock prices and market trust. 

The role of assurance in XBRL reporting 

XBRL reports are currently signed off by company senior management and assured by auditors, ensuring that the digital data provided is reliable for users. AI can assist in mapping financial and non-financial data to the correct XBRL tags, but it cannot replace the assurance process that guarantees digital data integrity and usability. 

Several AI-based XBRL-tagging tools are already on the market, and they work reasonably well. However, they still make mistakes, especially when dealing with complex, unique reporting needs or newly introduced regulations. Since AI can only learn from past data, it may struggle to apply new financial or non-financial reporting standards. Some of the errors these tools make are significant, requiring companies and auditors to validate and correct the automated tagging. 

However, companies can complete the tagging process much faster and more accurately by leveraging AI-based XBRL-tagging tools. If a company is still manually tagging XBRL reports, then it’s understandable that they may find XBRL reporting cumbersome and ask for simplification—but perhaps it’s time to modernize these companies’ approach. 

XBRL combined with AI is the future 

Another key consideration is that regulators, auditors, investors, and other capital providers demand traceable, explainable reporting. AI models, particularly complex deep learning algorithms, often function as a “black box”—making it difficult to explain how they arrive at certain conclusions. 

Furthermore, AI can assist in detecting errors and anomalies for companies, auditors, and regulators—but it cannot substitute for XBRL’s rule-based, structured approach. 

Conclusion: Keep the XBRL requirement—AI can support, not replace It 

We Mean Business Coalition strongly advocates for maintaining the requirement for XBRL reporting—both for financial and non-financial disclosures—to ensure access to high-quality, structured data that is suitable for analysis and freely available to all stakeholders: retail and institutional investors, other capital providers, business authorities and regulators, academics, journalists, and other researchers.

Read our new paper on data quality  to learn more about the root causes of ESG data challenges for data users and discover actionable solutions from companies, data providers, data users and legislators to enhance reliability and usability. Do also see the appendix, if you want tips and tricks as a data user, to identify data quality issues.