A US-based researcher, Ayobami Gabriel Olanrewaju, has developed a new artificial intelligence algorithm that has the potential to revolutionise how investment portfolios adapt to rapidly changing market conditions.

This system uses reinforcement learning, a machine learning method that learns by trial and error, to adjust asset allocations in real time based on live price movements and trading volumes.

Unlike traditional models that rebalance at fixed intervals, this algorithm continuously applies value-at-risk thresholds (limits on potential losses) and factors in transaction costs each time the portfolio is adjusted.

In simulated trials using data from the Standard & Poor’s 500 Index, the model showed a five per cent improvement in risk-adjusted returns compared to traditional strategies. It moved funds to lower-volatility assets during sharp declines and then reallocated to higher-growth opportunities as conditions improved, all without manual intervention.

A companion study, published in the International Journal of Research in Finance and Management, tested the algorithm across stocks, bonds, and commodities under various stress scenarios. The results confirmed that the system maintains its performance even during market sell-offs or periods of low trading volume, proving its versatility across different types of investments.

According to Ayobami, the algorithm’s design also automates compliance checks, ensuring regulatory risk limits are enforced during each decision cycle. “By building risk controls into the learning process, we eliminate downstream overrides and create a clear, auditable record of every trade decision,” he states. This approach simplifies operations and helps portfolio managers meet oversight requirements more efficiently.

Looking ahead, Ayobami reveals he is working to expand the model to cover multi-asset portfolios and improve its speed, aiming for decision-making within a second. He also added that he is developing a user interface to translate the algorithm’s complex signals into clear dashboards, enabling investment teams to monitor and manage adaptive allocations with ease.

As global markets become increasingly volatile, this next-generation model offers a scalable blueprint for intelligent trading systems that combine real-time adaptability, built-in compliance, and enterprise-level reliability, setting a new standard for resilient portfolio management.