MV Capital - Central Bankers Raise Alarms Over Agentic AI Risks in Financial Markets

European officials warn that agentic AI systems could amplify market volatility and outpace traditional regulation, calling for new safeguards amid rapid technological change

Agentic AI and Market Stability Concerns

Central bankers from Europe have voiced serious worries about agentic AI, which refers to systems capable of autonomous actions toward set goals with minimal human input. These officials argue that such technology introduces new challenges for financial systems, particularly during periods of market pressure.

At the European Central Bank’s annual meeting in Sintra, Portugal, Bank of England deputy governor Sarah Breeden addressed the potential for these systems to increase instability. She raised questions about whether mechanisms similar to circuit breakers or kill switches should be put in place to limit or halt trading if faulty AI actions lead to widespread disruption.

Breeden stated that agentic AI could worsen volatility when markets face strain, suggesting operational limits may be necessary to prevent localized issues from spreading.

Regulatory Timelines Lag Behind AI Development

Officials noted that standard regulatory processes, which involve extended consultation and implementation periods, do not align with the quick pace of AI advancements that can occur over weeks or months. This gap creates difficulties in managing emerging risks in real time.

UK Financial Conduct Authority CEO Nikhil Rathi discussed this mismatch in comments to CNBC. He pointed out that regulators require fresh approaches and closer cooperation with market participants rather than relying on outdated cycles.

“Some of these technologies now move in weeks or months, and the traditional cycle of rulemaking simply doesn’t work in that way, so we need to think about new tools and a different way of working with the market in a more collaborative way.” Nikhil Rathi, CEO of the UK’s Financial Conduct Authority

Cybersecurity Risks Amplified by AI Advances

European Central Bank President Christine Lagarde described AI as presenting a major risk, particularly in the area of cybersecurity. She noted that while past concerns focused on hacking and data issues, current AI capabilities accelerate threats in ways that existing defenses struggle to address.

“For about a decade now, we have been talking about cybersecurity risks, hacking, data theft, and so on. But with the acceleration and deepening of AI models, we are confronted with a much more serious risk, because it is happening very, very quickly, and because the means of defense — and the funding required for them — have yet to be found.” Christine Lagarde, President of the European Central Bank

Lagarde’s comments draw attention to the need for stronger resources to match the speed of AI progress.

Financial Stability and Economic Feedback Risks

Beyond immediate operational issues, institutions like the Bank for International Settlements have flagged broader stability concerns. In late June, the BIS cautioned that excessive enthusiasm around AI could result in sharp corrections in related asset prices, especially if monetary policy tightens to address inflation. Such shifts could generate disruptive macro-financial feedback loops.

Breeden also referenced rising debt levels tied to AI investments, noting that any decline in asset values could carry greater stability implications.

Key points from central bank discussions:Agentic systems may create correlated behaviors across participants, leading to amplified price movements under stress.
Debt financing structures could transmit shocks more effectively during corrections.
Maturity mismatches between long-term AI infrastructure and shorter-term funding add pressure points.

IMF Monetary and Capital Markets Department director Tobias Adrian highlighted potential mismatches in asset and debt durations, which could complicate refinancing when conditions change.

Calls for Practical Safeguards

European policymakers are pushing for guardrails that can adapt to fast-moving developments. Discussions include technical standards, ongoing monitoring, and dynamic responses tailored to AI behaviors in live markets.

Breeden’s remarks at the ECB event underscore the view that static rules at deployment may not suffice, with a need for tools that activate based on real-time conditions.

Institutions involved in the warnings include:Bank of England
European Central Bank
UK Financial Conduct Authority
Bank for International Settlements
International Monetary Fund