Samer Choucair: Explainable AI Is Becoming a Requirement for Risk Pricing
Investment entrepreneur Samer Choucair said the explainability of artificial intelligence decisions is shifting in 2026 from a technical advantage to a fundamental requirement for risk pricing and capital reallocation, as AI moves beyond experimentation and into operational deployment across banks, financial institutions, insurers, e-commerce companies, and regulated industries.
According to Samer Choucair, the investment question is no longer limited to how accurately an AI model can predict an outcome. It increasingly centers on whether an institution can explain a decision and trace its path from the underlying data to the model and ultimately to the action taken.
“An institution may deploy an advanced model, but if it cannot defend the resulting decisions before an auditor, regulator, or customer, it may possess sophisticated technology without possessing a scalable operational asset,” Choucair said.
Choucair explained that comments from FICO executives in August 2026 have renewed attention around what can be described as the decision layer: the infrastructure connecting data, models, rules, governance, monitoring, and ultimately the execution of a decision.
He believes this layer will become increasingly important in the valuation of fintech and enterprise-software companies because it has the potential to transform governance from a compliance expense into an operational asset capable of supporting recurring revenue.
Samer Choucair noted that global spending on artificial intelligence continues to increase, while operational returns remain uneven across institutions. Within regulated industries, the consequences of a poor AI-driven decision cannot be measured solely by a decline in statistical accuracy. Such decisions can create legal, regulatory, and operational liabilities.
For that reason, the ability to reconstruct the entire decision chain — from data and model inputs through to the final action — is becoming an increasingly important component of the quality and value of the technology itself.
Choucair said the European Union’s AI Act reflects this transition through stricter requirements governing high-risk AI systems, including certain applications used to assess creditworthiness. The regulatory framework reinforces the importance of risk management, transparency, documentation, and human oversight, while similar standards could increasingly influence other markets through global banks and multinational institutions.
In Saudi Arabia and the wider Gulf, these developments are directly connected to the region’s digital-transformation agenda and Saudi Vision 2030, particularly as artificial intelligence becomes more widely deployed across credit, payments, fraud detection, insurance, and government services.
Choucair argued that investment at this stage should therefore focus not simply on the size or sophistication of AI models, but on the infrastructure that makes artificial intelligence operational, measurable, governable, and scalable.
According to Samer Choucair, institutional investors evaluating AI companies should ask three fundamental questions: Does the company possess a genuine decision layer, or is it simply selling a model? Can its technology operate effectively within regulated industries? And can spending on governance and compliance ultimately be converted into recurring revenue and measurable economic value?
Choucair believes some of the most significant opportunities could emerge among companies specializing in decision software, AI governance, and risk management, as well as banks capable of using artificial intelligence to expand credit and reduce the cost of risk without simultaneously increasing their regulatory liabilities.
The investment implications extend beyond identifying which company has developed the most powerful model. As AI becomes embedded in increasingly consequential financial and commercial decisions, the ability to demonstrate how those decisions were reached could become a competitive differentiator in its own right.
Samer Choucair concluded that the next phase of the artificial intelligence race will not be defined solely by who builds the largest or most sophisticated model. It will increasingly favor institutions capable of converting prediction into governed, auditable, and accountable decisions.
“The most accurate model is not necessarily the most monetizable asset,” Choucair said. “Real value begins when a decision can be understood, reviewed, and converted into sustainable economic value.”