AI Adoption and Indirect Tax Compliance: The EU's Emerging Governance Crisis
A survey of 176 tax leaders reveals a critical gap between AI adoption and audit defensibility in indirect tax compliance, coinciding with the EU's intensified enforcement under the VAT in the Digital Age (ViDA) initiative. The findings expose structural weaknesses in governance frameworks, placing organizations at heightened risk during a period of increased tax authority scrutiny.
Context
The survey, conducted in Q2 2026, captures a moment of rapid AI adoption within indirect tax functions across the EU. With 92% of organizations utilizing AI in some capacity, the technology's role in tax compliance is undeniable. However, this adoption has outpaced the development of necessary governance structures, creating significant compliance risks.
Key stakeholders include tax authorities implementing ViDA provisions, corporate tax functions under pressure to adopt AI, and compliance technology providers. The findings are particularly relevant as EU member states finalize their real-time reporting infrastructures under ViDA, which includes mandatory e-invoicing and continuous transaction controls.
The Audit Defensibility Gap
The most pressing concern identified is the audit defensibility gap: 57.4% of respondents expressed low confidence in their ability to defend AI-assisted tax decisions during an audit. This majority posture indicates systemic vulnerabilities, especially as 88.6% of tax functions report active pressure from leadership to expand AI use.
Shadow Adoption and Governance Vacuum
The governance crisis is structural. Only one-third of organizations have formal sign-off processes for AI outputs, while 46.6% acknowledge their current processes would fail external scrutiny. The dominant adoption model is "shadow adoption," where 60.8% of organizations rely on individuals using public AI tools without established governance or audit trails.
This informal approach is compounded by vague mandates: 76.1% of teams operate under broad "do AI" directives, with only 12.5% receiving specific targets or KPIs for adoption.
Operational Realities
Despite widespread AI adoption, operational efficiency gains remain elusive. Over the past two years, 85.8% of teams experienced increased workloads, while only 31.2% saw headcount growth. Just 6.8% of organizations reported decreased compliance workloads over the past year, and 71% have not fully automated a single indirect tax workflow end-to-end.
The High-Performing Outlier
A notable exception exists: 2.8% of organizations have fully formalized AI accountability structures. These teams experience compliance workload reductions at eight times the rate of their peers, demonstrating that robust governance—not mere AI adoption—drives efficiency gains.
Implications for Tax Functions
The findings compel tax functions to prioritize governance frameworks alongside AI adoption. Immediate steps include:
- Formalizing Sign-Off Processes for all AI-generated outputs, ensuring traceability and defensibility.
- Documenting Decision-Making through clear audit trails that connect AI inputs, processes, and outputs.
- Establishing Clear KPIs tied to AI adoption that balance innovation with compliance risk management.
- Phasing Out Shadow Adoption by integrating decentralized AI use into centralized, governed workflows.
Organizations failing to address these gaps risk heightened enforcement actions under ViDA, particularly as tax authorities increase real-time transaction monitoring.
Outlook and Next Steps
The survey highlights an urgent need for standardized AI governance frameworks within indirect tax compliance. Tax functions should monitor developments in:
- ViDA Implementation Timelines: As member states finalize their real-time reporting systems, compliance expectations will evolve.
- Tax Authority Guidance: Emerging best practices from national tax administrations on AI defensibility in audits.
- Industry Collaboration: Initiatives to develop cross-border governance standards for AI in tax.
The 2.8% of organizations that have closed the defensibility gap offer a roadmap for others, demonstrating that proactive governance can transform AI from a compliance risk into an operational advantage.
