Beyond proof of concept: a Riskfuel surrogate of BBVA's autocallable pricing model now supports pre-trade structuring in the bank's ePricer platform

TORONTO, Nov. 10, 2026 /CNW/ -- Riskfuel, a financial technology company that uses AI to accelerate the pricing of complex derivatives, today announced that BBVA is using a Riskfuel neural surrogate in the pre-trade structuring of autocallable products, as part of the bank's ePricer platform for structured products. The deployment moves the technology beyond proof of concept and into the workflow structuring teams use to design trades for clients.

The surrogate is a deep neural network trained on data generated by BBVA's own pricing model, and the significance of the work lies in how completely it reproduces that model on real trades. It takes the same inputs as the model it learns from: more than 4,000 input features spanning full volatility surfaces, interest rate curves and discrete dividend schedules, together with structural features such as autocall, coupon and knock-in barrier levels. The surrogate was measured against the bank's reference model across more than 100,000 test trades, spanning maturities of up to five years, and its prices closely tracked the reference across that domain.

While the work has taken domain coverage and accuracy to a new level, it is the performance gain that changes the workflow. Benchmarks measured by Riskfuel and Intel on Intel® Xeon® 6 processors with Intel® Advanced Matrix Extensions (Intel® AMX) demonstrated approximately 2.27 million valuations per second.

"What is important here is the combination of speed and consistency with our own pricing models. By integrating Riskfuel's technology into ePricer, we can dramatically accelerate the evaluation of different autocallable structures while keeping BBVA's pricing framework at the core of the process. That gives our structuring teams a much more efficient way to explore and compare alternatives."

— Francisco Gómez Casanova, Executive Director, XVA Quantitative Analysis, BBVA Corporate & Investment Banking

This combination of speed and accuracy unlocks a better approach to structuring autocallables. Designing an autocallable involves comparing a very large number of possible combinations. Because autocallables require computationally intensive stochastic simulation to value correctly, desks have had to work from a shortlist of promising ideas. With the surrogate, a much broader range of candidate structures can be priced and ranked quickly, giving structuring teams a more efficient way to explore and compare alternatives. Evaluating thousands of candidate trades, a job which previously would have taken hours, now takes less than a minute.

"The hard part was never about making a neural network fast. It's about making one that a bank can actually use," said Ryan Ferguson, CEO of Riskfuel. "That means reproducing the institution's own pricing model, across the real trades on its books, at the accuracy its own people expect. Getting there is what turns this from an interesting result into a practical tool for a desk's day-to-day structuring workflow."

Because the surrogate is a neural network, it runs on the general-purpose Intel Xeon processors already used to run quantitative pricing libraries, using the matrix acceleration built into the processor. No specialized accelerator hardware, proprietary framework or separate AI operating model is required.

About Riskfuel
Riskfuel builds neural-network surrogate pricing models for derivatives, reproducing an institution's own pricing models at a fraction of the computational cost. Riskfuel is headquartered in Toronto. For more information, visit riskfuel.com.

About BBVA
BBVA is a global financial group with a customer-centric vision, characterized by its pioneering strategy in digitalization, innovation and sustainability. BBVA holds a leading position in Spain, is the largest financial institution in Mexico, and has leading franchises in South America and Turkey. In Europe, BBVA has a steadily growing presence, driven by its specialized offices serving business customers and its focus on digital banks, currently operating in Italy and Germany. In addition, BBVA provides specialized services to large corporations through its offices in the United States and Asia. This strong geographic diversification, together with high levels of customer acquisition through digital channels, positions us to strengthen our leadership and address the challenges of the future.

Its division BBVA Corporate & Investment Banking (BBVA CIB) brings together the activities of investment banking, markets, financing and transactional services for institutional investors and corporate clients. It has a strong global presence, providing services in 25 countries through an extensive team of experts, including investment banking specialists and advisors in specific industries and sectors. BBVA CIB offers a wide range of value-added products and financial solutions, for the simplest needs and for the most complex ones. Its mission is to help clients to carry out their projects and achieve their business, transformation and sustainability objectives, whether they are local or international.

SOURCE Riskfuel Analytics Inc.