Image Credit: Unsplash

Vertus, a cognitive artificial intelligence company, has independently audited results that verify that it delivered a return of just over 51 percent for the 2025 fiscal year.

According to the same audit, its Sharpe ratio, a common measure of risk-adjusted performance, was 2.13.

The audit also confirms that 11 months during the year were positive, with a maximum drawdown of around 10 percent that was recovered very quickly. The audit also shows that the company’s daily trading volume many times exceeded the billion-dollar per day mark.

According to the company, the difference comes from how most AI-driven investment systems are designed. The reason is that most AI-driven investment systems rely on pattern-recognition architectures that are trained solely on historical data, a design flaw that produced significant difficulties when they tried to adapt to conditions outside of their prior training and that had never been seen before.

Vertus co-founder Julius Franck described the company’s approach as fundamentally different and starting at the architectural level. Rather than deploying a fixed model trained on historical data, the Vertus system has an inherent flexibility that constructs a new adaptive and reasoning structure for each problem it encounters. A real-time awareness of changing conditions is integrated into the reasoning process from the outset rather than retrieved from static and older datasets. According to the company, the system adapts its underlying models as conditions change, rather than relying on assumptions that may no longer apply.

Vertus doesn’t operate as a fund. The company functions as the cognitive and execution infrastructure for partner funds that deploy its reasoning architecture in live markets. The company was co-founded by Franck, Alex Foster, and Michal Prywata, whose combined backgrounds span quantitative finance, algorithmic trading infrastructure, medical robotics, aerospace engineering, and agricultural biotechnology.

Vertus has made its cognitive reasoning architecture accessible to external developers and institutions through a live API covering applications in healthcare, supply chain management, legal services, and scientific research.

The information provided in this article is for general informational and educational purposes only. It is not intended as financial or professional advice. Readers should not rely solely on the content of this article and are encouraged to seek professional advice tailored to their specific circumstances. We disclaim any liability for any loss or damage arising directly or indirectly from the use of, or reliance on, the information presented.

Disclaimer: GeekWire newsroom and editorial staff were not involved in the creation of this content..