News Summary
In 2026, the UK Financial Conduct Authority (FCA) continued to advance its ambition of becoming a "smarter regulator", expanding its use of artificial intelligence and advanced data analytics across supervisory functions. The regulator has been exploring AI-enabled tools to support areas including authorisation processes, risk identification and the allocation of supervisory resources.
Similar work is underway internationally. Central banks, financial supervisors and international standard-setting bodies, including the Bank for International Settlements (BIS) and the Financial Stability Board (FSB), have begun exploring how AI can strengthen regulatory capacity. In June 2026, the FSB published its consultation on Sound Practices for the Responsible Adoption of Artificial Intelligence, signalling that AI is gradually becoming part of the infrastructure through which supervision itself is conducted.
Supervisory Tool
AI is becoming part of the workbench regulators use to observe complex financial markets.
Financial supervisors operate in an environment defined by information complexity. They review regulatory filings, monitor market activity, analyse consumer complaints and assess risks across thousands of institutions. The volume and speed of financial data have expanded considerably, while traditional supervisory methods still depend heavily on human review and institutional experience.
Every supervisory decision begins with information. Reports must be read, complaints assessed and unusual patterns identified before regulatory judgement can even begin. As these tasks become data-intensive, AI is beginning to change that process. Supervisors are adopting machine learning technology to identify market abuses, analyze natural language in corporate disclosures, and dynamically allocate regulatory resources. The purpose of these tools is not to replace regulatory judgement. Their value lies in extending the ability of supervisors to observe, analyse and respond to increasingly complex markets. As BIS points out, regulators are not only investing resources in Artificial Intelligence tools, but are also working to build the institutional capabilities needed to verify and supervise AI systems.
Supervisory Capacity
The pressure is not novelty; it is information scale.
A payment failure in one institution may now affect firms operating across several jurisdictions. Trading data arrive continuously rather than at fixed reporting intervals. Supervisors therefore face more information than manual processes were designed to handle. The FCA has also emphasised the role of data-driven supervision as part of its regulatory approach. By using advanced analytical tools, regulators can move beyond periodic reviews and develop a more continuous understanding of market behavior.
Similar developments can be observed elsewhere. European supervisory authorities have long relied on data analytics to support financial stability monitoring and market oversight. In the United States, agencies such as the Securities and Exchange Commission (SEC) have used technological tools to analyse trading activity and identify potential market abuse. Across jurisdictions, AI is becoming part of the toolkit through which regulators understand financial markets. In recent work on AI and financial supervision, the BIS notes that supervisory authorities are investing not only in AI applications, but also in the institutional capabilities required to understand, validate and oversee AI-enabled systems. As supervisory technologies become more sophisticated, the challenge extends beyond technical adoption to ensuring that supervisory expertise develops at the same pace.
By analysing transaction patterns, corporate disclosures and market behavior, AI allows supervisors to compare millions of transactions, detect changes that might otherwise pass unnoticed and prioritise cases requiring closer examination, which can further highlight potential connections between entities, detect changes from historical patterns and help regulators decide where closer examinations may be needed.
Yet, if supervisors rely on AI-generated insights, they must understand how those systems operate, what data they depend on and how their outputs influence regulatory decisions.
Shared Concerns
Regulators using AI face the same accountability logic they expect from firms.
When a bank uses AI for credit assessment, supervisors expect the institution to understand its model, maintain appropriate controls and remain accountable for decisions affecting customers. The same logic applies when regulators use AI. Supervisory decisions influence market participants, consumer outcomes and financial stability. Any AI-assisted judgement therefore requires appropriate safeguards around transparency, human oversight and institutional accountability. From the perspective of compliance and management, automated regulatory alerts or risk scores must remain explainable, transparent, and legally defensible.
By introducing risk-based categories and common governance requirements, the EU framework establishes expectations around areas such as risk management, transparency and human oversight for high-risk AI systems. These principles provide a reference point for financial supervisors considering how AI should be governed within regulated environments. The FCA has chosen to rely on existing regulatory structures, including governance requirements and senior management responsibilities, rather than introducing a dedicated financial AI law. The underlying assumption is firms should remain accountable for how they deploy technology, regardless of whether decisions involve traditional models or newer AI systems.
Although the regulatory paths differ, expanding supervisory capability doesn't diminish institutional responsibility. In 2026 consultation on Sound Practices for the Responsible Adoption of Artificial Intelligence, the FSB emphasises that effective AI governance depends not only on technological performance, but also on governance arrangements that preserve accountability, human oversight and ongoing monitoring throughout the lifecycle of AI systems. This suggests that, whether it is a financial institution or a regulatory agency, maintaining this accountability requires solid data and model provenance to ensure that each AI-assisted judgment can be independently audited and traced.
To sum up, the future of financial regulation will depend not only on how effectively firms deploy AI, but also on how well regulators understand and supervise AI-enabled markets. In this sense, AI is gradually becoming part of the infrastructure of financial governance itself.
Looking Ahead
Supervisory AI will be judged by provenance, oversight, and institutional capacity.
From the EU's attempt to establish common AI rules, to the UK's emphasis on institutional responsibility, and now regulators adopting AI themselves, financial governance has evolved in an era shaped by AI. To uphold market trust and regulatory integrity, institutions must ensure they possess the technical resources, organizational capabilities, and human judgment required to manage the technologies they deploy.
