The European Institute of Management (EIM) is partnering with Perpetuals.com to create a joint research group focused on artificial intelligence in financial markets. More than 20 doctoral researchers will work with Perpetuals executives and access proprietary trading datasets and live venue data to study behavioral finance, market microstructure, trader classification and model risk.

Artificial intelligence is moving deeper into financial-market infrastructure, and a new academic-industry research partnership aims to examine what happens when machine-learning systems begin influencing trading decisions at scale.

The European Institute of Management has launched a joint research group with Perpetuals.com Ltd. (Nasdaq: PDC), bringing more than 20 doctoral researchers into collaboration with the AI-focused financial technology company.

The initiative is expected to operate for at least three years, broadly corresponding to the doctoral research cycle. Researchers will work alongside Perpetuals leadership and its data-science and engineering teams, using proprietary trading datasets and live venue data to investigate several areas of quantitative finance.

Those areas include behavioral finance, trader-performance classification, market microstructure and model risk.

The combination is significant because financial AI is moving beyond forecasting exercises conducted on historical datasets. Trading systems increasingly operate within live markets where their own actions can influence liquidity, price formation and the behavior of other participants.

A 2026 theoretical study published in Finance Research Open argues that AI can change market microstructure by accelerating information processing and altering liquidity formation, volatility and price discovery. The research also highlights the possibility of temporary distortions as competing AI strategies adapt to one another.

That makes the research agenda announced by EIM and Perpetuals particularly relevant.

Rather than focusing exclusively on whether an AI model predicts prices accurately, researchers can examine how models interact with the market itself. A strategy may appear effective in a historical backtest but behave differently when market conditions change, when other algorithms respond to the same signals or when liquidity becomes constrained.

Model risk is consequently becoming a central issue for financial AI.

The challenge extends to behavioral finance. Trading algorithms may be designed to remove human emotion from decision-making, but the data used to train them contains patterns created by human participants. An AI system can therefore learn relationships that reflect investor biases, momentum behavior or other recurring market dynamics.

Recent academic research has also begun examining the behavior of autonomous AI agents in simulated financial markets. A 2026 study by researchers from Oxford’s Saïd Business School and the Chinese University of Hong Kong, Shenzhen found that large language model-based trading agents can display behavioral patterns resembling established investor biases and can contribute to bubble-like market dynamics in experimental settings.

The EIM-Perpetuals research group is positioned to explore similar questions using real-world market data.

For Perpetuals, the partnership also provides access to a larger pool of quantitative research talent. More than half of the participating researchers are already employees of the company, according to the announcement, meaning the initiative will combine doctoral research with an existing engineering and data-science organization.

Perpetuals has been building an infrastructure stack that combines AI, financial-market technology and blockchain. The company describes its Kronos X platform as a regulated trading infrastructure supporting digital assets, derivatives, tokenized assets and structured products, while its machine-learning systems are designed to analyze trade-level risk.

That positioning gives the research partnership a broader context.

Financial institutions have traditionally separated academic quantitative research from production trading infrastructure. The emergence of AI is narrowing that divide. Models now require continuous testing, monitoring and adaptation, while research questions can emerge directly from production environments.

The arrangement also reflects a broader fintech trend toward AI-native market infrastructure.

Trading firms and exchanges are investing in machine learning for execution, risk management, surveillance, pricing and liquidity analysis. At the same time, blockchain-based financial infrastructure is creating new trading venues and settlement models. Companies operating across both areas face the additional challenge of understanding how AI behaves in markets that may operate around the clock.

Perpetuals has already positioned itself around that convergence. Following its combination with EarlyWorks in January 2026, the company described its platform as combining blockchain infrastructure with AI-driven trading technology and targeting regulated digital-asset and derivatives markets.

For financial institutions, the research has implications beyond trading performance.

As AI systems become more autonomous, institutions need to understand not only whether a model produces profitable signals but also how it behaves under stress, whether its outputs can be explained, how correlated multiple AI strategies become and whether models create new systemic risks.

A September 2026 study in the Journal of Behavioral and Experimental Finance found that higher AI penetration in an agent-based market model reduced ordinary volatility while increasing similarity among AI strategies and sensitivity to common shocks. The result suggests that lower day-to-day volatility does not necessarily mean lower overall market risk.

That is precisely the type of distinction that model-risk research can help clarify.

The partnership could therefore produce value on two levels. For Perpetuals, doctoral researchers provide additional expertise for developing and testing financial AI systems. For EIM, access to real trading infrastructure provides researchers with an environment in which theoretical models can be tested against operational market conditions.

There are limits, however. Proprietary trading data is not automatically equivalent to a representative view of global markets, and results derived from one trading environment may not generalize to other asset classes or venues. The quality of any resulting AI models will also depend on data governance, validation methodology and the separation between research objectives and commercial incentives.

Those considerations make independent validation important.

For the broader fintech ecosystem, the collaboration nevertheless illustrates how financial AI is evolving. The next generation of trading technology will likely require expertise spanning machine learning, quantitative finance, behavioral economics, market structure, cybersecurity and regulatory risk.

Academic-industry partnerships may become one way to build that multidisciplinary capability.

The larger question is no longer simply whether AI can trade. It is how AI changes the market in which it trades—and whether financial institutions can measure those effects before they become operational or systemic problems.

Market Landscape

AI-driven trading is becoming a significant area of financial technology infrastructure, with machine learning increasingly applied to forecasting, execution, risk management, surveillance and portfolio construction.

The competitive landscape includes quantitative hedge funds, banks, exchanges, fintech startups and specialist trading-technology providers. As AI systems become more capable, the focus is shifting from predictive accuracy alone toward market microstructure, model risk, explainability, strategy correlation and financial stability.

The EIM-Perpetuals initiative is notable because it combines academic research with access to proprietary and live-market data. That approach could help researchers study AI behavior under real trading conditions rather than relying exclusively on simulations or historical datasets.

For GlobalFinTechEdge, the development is especially relevant to AI-powered trading, blockchain financial technology, digital-asset infrastructure, quantitative finance, market intelligence and banking technology innovation.

Top Insights

• EIM and Perpetuals are combining doctoral research with real trading data to investigate how AI affects financial-market behavior and risk.
• More than 20 doctoral researchers will study behavioral finance, trader classification, market microstructure and model risk over an expected three-year initiative.
• Academic research increasingly suggests AI can alter liquidity, price discovery, volatility and strategy correlation rather than simply improve prediction accuracy.
• Perpetuals is positioning the initiative alongside its broader AI, blockchain and regulated financial-market infrastructure strategy.
• The partnership highlights growing demand for multidisciplinary expertise spanning machine learning, quantitative finance, market structure and financial risk.

Get in touch with our fintech expert