The Bank of Italy has been running a joint project with the OECD and the European Commission since early 2025, aimed at developing a comprehensive framework for AI in Italian financial markets. The project concluded its main analytical phase in spring 2026 and Deputy Governor Chiara Scotti presented the findings at a conference on strengthening the regulatory and supervisory framework for AI in Italian finance – a speech she also gave in the context of broader concerns about AI governance in the global financial system. YourDailyAnalysis unpacks why a mid-sized central bank engaging this deeply with global AI governance matters more than it might initially appear.
The substance of Banca d’Italia’s work spans four areas: improving the regulatory framework, strengthening supervisory approaches, promoting innovation through regulatory sandboxes and innovation facilitators, and increasing the efficiency of authorities’ own internal processes through AI deployment. That last element is underappreciated. Central banks are not merely regulating AI in finance – they are adopting it themselves, as risk management tools, as supervisory scanning instruments, and as data processing infrastructure. The OECD’s May 2026 report on AI supervision in finance, which surveyed practices across member countries, provides the empirical backbone for Italy’s policy review.
The EU AI Act, which entered into force in August 2024 and whose most significant requirements become applicable from August 2026, represents the compliance deadline that concentrates minds at Banca d’Italia and peer institutions. The Act takes a risk-based approach, classifying AI systems from minimal risk to prohibited, with financial services applications clustered in the high-risk category where obligations are most onerous. Italy has been ahead of most EU member states in using AI for supervisory purposes: Banca d’Italia adopted the first AI regulation by any Italian public authority in March 2022, covering the use of AI for complaints management. That early mover status gives its policy views credibility at the ECB and FSB level.
What makes the Bank of Italy’s engagement with global AI governance genuinely significant is the cross-jurisdictional dimension. YourDailyAnalysis identifies the OECD-Italy project’s private sector dialogue component as a model for how regulators should engage with AI firms before enforcement cycles begin – something the EU’s own AI Act process handled less effectively. The OECD project explicitly engaged private sector participants across Italy and other advanced economies, combining structured dialogue with evidence-based analysis. That process – rather than the eventual policy recommendations – is the real output: it builds a common language between regulators and the firms they regulate before the AI Act’s hard requirements hit. Editors at YourDailyAnalysis interpret Panetta’s broader positioning on global economic fragmentation, including his May 2026 speech at London’s “Cross-Border Payments at a Turning Point” event, as consistent with Banca d’Italia’s institutional instinct to use multilateral forums to shape rules before those rules are written by others.
There is a counter-argument that frames European central bank engagement with AI governance as elaborate process-management that leaves the actual AI development – the models, the compute, the frontier capabilities – entirely in the hands of a handful of U.S. and Chinese firms. That critique has force. The EU AI Act and the Bank of Italy’s supervisory framework govern how AI is used in European financial markets, not who builds the models. European regulators are writing the rules for a car they did not manufacture and cannot redesign. The practical implication is that European banks and asset managers will spend compliance resources on governance processes while their counterparts elsewhere compete on AI capability. That asymmetry is real and growing.
The May 2026 deadline concentration – when the most demanding EU AI Act requirements activate, when the OECD-Italy project concludes, and when financial institutions face their first substantive compliance reviews – makes the next six months the clearest test of whether European AI governance has substance or is sophisticated paperwork. The reporters at Your Daily Analysis will track whether any financial institution receives the EU AI Act’s first enforcement action in this window, which would reveal more about real regulatory intent than any policy speech.
