In a stark warning about the systemic risks posed by generative artificial intelligence in financial markets, European Stability Mechanism economists Thiago Fauvrelle and Martin Rohár argued that complex financial innovations may outpace regulation, leaving markets vulnerable to unknown shocks. In a detailed blog post published on 28 July 2025, they called for urgent coordination and new risk frameworks to manage AI-driven complexities across European markets.
“Large language models capable of generating human-like text have evolved rapidly over the past few years. With this leap in technology, it became possible to generate coherent text using artificial intelligence (AI) which, in turn, led to the proliferation of so-called generative AI (GenAI). Massive improvements in extracting insights from extensive domain-specific knowledge brought about by GenAI is widely expected to reshape parts of the economy that rely on processing large amounts of information, like the financial sector,” they wrote.
Model complexity
The financial industry, they noted, is inherently dependent on “continuous aggregation, analysis and exchange of data”, and has historically been quick to integrate new technology--from the adoption of mainframe computers to the rise of electronic trading platforms. However, GenAI is different, they said, because it “substantially improves both generation and synthesisation of qualitative information,” adding a new level of complexity that could obscure emerging systemic risks.
“Technological advancements in finance underscore a somewhat paradoxical truth: the very tools that improve the efficiency of processes [are what] make the system more complex. Increased system complexity creates unpredictable interdependencies and could increase potential failure points,” they warned.
Retail investors
One of the most immediate concerns, according to the economists, relates to retail investors. GenAI-powered platforms could “help overcome” market entry barriers and product complexity by providing “personalised investment recommendations” and “facilitating seamless execution of investment and trading strategies using natural language input.” While this could “improve financial inclusion” and “enhance market efficiency,” it also introduces “new sources of volatility and systemic risk.”
A particularly concerning behaviour, they wrote, is “vibe trading”--a scenario where retail investors rely on GenAI prompts to generate investment strategies and then execute them “without thorough critical evaluation, deferring largely to the GenAI’s implicit inference.” This could lead to “correlated strategies derived from similar model training data,” increasing the risk of volatility and liquidity shocks.
“Retail investors may be more prone to panic selling or speculative behaviour during market stress, lacking the sophisticated risk management and dynamic investment strategies, and regulatory oversight of institutional investors,” they argued.
Autonomous trading systems
Beyond the retail segment, Fauvrelle and Rohár also raise concerns about the rise of fully autonomous AI investment systems, particularly in institutional settings. These systems, unlike traditional algorithmic models that operate with fixed rules and human oversight, are evolving toward greater independence and adaptability.
“Some institutional investors are beginning to deploy AI-enabled systems capable of making investment decisions without human oversight,” they noted. Unlike earlier algorithmic approaches, “new systems that integrate cutting edge AI algorithms can adapt their strategies based on market conditions and learn from their outcomes without human oversight.”
This shift, the authors suggest, introduces new uncertainty. While such adaptability might enhance performance in stable conditions, they caution that it “makes it extremely difficult to predict how these systems will behave during extreme market events or stress conditions where historical data may provide limited guidance.”
Lessons from past crises
Fauvrelle and Rohár drew historical parallels to show how financial innovation can sow the seeds of instability if not properly understood or supervised.
“The 1987 stock market crash was exacerbated by portfolio insurance programmes that automatically triggered selling based on market conditions. The 1998 collapse of long-term capital management illustrated how the failure of complex quantitative models could rapidly become systemic threats. Similarly, the widespread use of mathematical models and sophisticated derivatives by the turn of 20th century promised better risk management, but ultimately contributed to the 2008 financial crisis when these tools were poorly understood until the system was under severe stress.”
They added that “new generations of AI models have the potential to create hidden vulnerabilities that only become apparent under stress.”
Call for regulatory overhaul
In light of these risks, the economists called for tighter monitoring and a wholesale rethink of how regulators assess financial stability threats in an AI-driven environment.
“The transformative impact of GenAI on financial markets demands an evolution in how financial stability risks are monitored, assessed and managed. The uneven pace and scope of GenAI integration across the financial sector call for proactive surveillance mechanisms that can identify emerging vulnerabilities before they manifest as systemic threat,” they stressed.
Coordination across financial institutions will be vital, they added, as the AI revolution continues to unfold. They concluded that the ESM would “improve its capabilities to stay ahead of the curve and preserve financial stability across its member states.”
