Product Manager

* Backbase * / Press * / Banking AI trained to admit uncertainty resolves more customer queries at a fraction of the cost. AMSTERDAM, 30th July, 2026: Backbase, the company behind the AI-native Banking OS, today released the findings of a peer-reviewed study on production-grade banking AI, presented at the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026). It was led by Denys Katerenchuk, Head of AI Research at Backbase, previously of Google and IBM. It is among the first peer-reviewed accounts of a banking-grade language model measured in live production. The hardest problem in customer-facing banking AI is what the model does when the evidence isn't there. A system that invents an answer about a fee, a rate, or a policy creates regulatory exposure. But one that declines too often becomes useless. The study shows this trade-off can be engineered out. A 12-billion-parameter model trained to recognize the limits of its own evidence resolved significantly more customer queries in live deployment. It also outperformed GPT-4.1 on the quality and grounding measures that matter most in a regulated environment, at a fraction of the cost. "Off-the-shelf models tend toward hallucination and sycophancy - confident, agreeable answers even without evidence. That's especially risky in banking, where information is complex, technical, and scattered across dozens of documents," explained Katerenchuk. The stakes of this disconnect are well documented. McKinsey estimates AI could drive up to 20% in net cost reductions for banks. Yet MIT research found that 95% of enterprise generative AI pilots deliver no measurable P&L impact. Katerenchuk and his team trained a model to understand the domain and recognize when information is incomplete. They taught it the boundaries of its own knowledge, so it says "I don't know" instead of inventing an answer. Key findings: * Honesty can be engineered: The model was trained on a dataset in which 22% of examples had no correct answer. That taught it the right response was an explicit "I don't know." It reached a 12% refusal rate, higher than the untuned base model's 4.3% and lower than GPT-4.1's 20.2%. The base model answered confidently even without evidence. GPT-4.1 declined questions it could have safely answered. * The honest model solved more customer problems: Over seven months at a large US financial institution, query resolution rose 7.1 percentage points across 3,297 sampled queries. That's a statistically significant gain. It came despite the model refusing nearly three times as often as its base. * A model a fraction of the size beat GPT-4.1: The model scored higher on independent evaluation: 6.21 versus 5.72 for GPT-4.1, on a 10-point scale. Citation grounding improved by 2.3 points, with answers cited directly to source documents. It also produced stronger results on FinanceBench, a public benchmark of SEC filing questions. * The economics undercut frontier models by an order of magnitude: Roughly $0.001 per query on a single GPU, 20-50x cheaper than GPT-4.1 and 3-5x faster. Training cost around $1,800. * Data order mattered more than data volume: On identical data, teaching general financial language first and calibrated refusal second produced the best model. Pooling everything at once collapsed answer quality by more than 40% and pushed refusals to nearly half of all queries. The findings land amid a live industry debate: research published by OpenAI in 2025 found that training and evaluation methods reward confident guessing over admitting uncertainty. The Backbase study offers production evidence of the alternative: a model rewarded for honesty, measured against real customers. "For three years, the AI industry has rewarded models for speed and confidence. Banking has rewarded itself for the same thing for three decades. Saying 'I don't know' got treated as a weakness, not a feature," said Jouk Pleiter, Founder and CEO of Backbase. "Our research shows the opposite: a model that knows the limits of its own evidence earns more trust, not less." "2026 is the year agentic workflows go live in regulated environments, but none of it works unless the model knows what it doesn't know," added Pleiter. This study is also the first published work from Backbase AI Research - the team that joined Backbase through its acquisition of Kasisto. The group focuses on the specific problems of AI in banking, publishing peer-reviewed research openly and moving findings directly into production. * The full paper, FinRAG-12B: A Production-Validated Recipe for Grounded Question Answering in Banking, was published in the Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Industry Track), July 2026 * The production analysis covers 3,297 randomly sampled customer queries over seven months (May-December 2025) at a large US retail credit union; institutions in the study are anonymised Backbase is on a mission to to put bankers back in the driver's seat. Backbase built the AI-native Banking OS - the operating system that turns fragmented banking operations into a Unified Frontline. Customers, employees, and AI agents work as one across digital channels, front-office, and operations. Backbase was founded in 2003 by Jouk Pleiter and is headquartered in Amsterdam, with teams across North America, Europe, the Middle East, Asia-Pacific, Africa and Latin America. 120+ leading banks run on Backbase across Retail, SMB & Commercial, Private Banking, and Wealth Management.

Backbase AI outperforms GPT-4.1 in banking by learning to say 'I don't know' Quick summary. Backbase has developed a 12-billion-parameter banking AI model that outperforms GPT-4.1 by admitting uncertainty. By training the system to say "I don't know" when evidence is missing, Backbase achieved a 7.1% increase in query resolution at 50x lower operational costs compared to frontier models. How does Backbase solve AI hallucinations in banking? Backbase solves the critical issue of AI hallucinations by training its 12B-parameter model, FinRAG-12B, to recognize the limits of its own knowledge. Unlike standard models that are rewarded for confident guessing, this system was trained on a dataset where 22% of examples had no correct answer, forcing the AI to prioritize grounded accuracy over sycophancy. * The model reached a 12% refusal rate, significantly more honest than the 4.3% rate of untuned base models. * It outperformed GPT-4.1 on citation grounding by 2.3 points, ensuring answers are tied to source documents. * Training costs were remarkably low at just $1,800, proving specialized models can beat general-purpose giants. What results has this AI delivered for financial institutions? In a seven-month live deployment at a large US financial institution, the model demonstrated that honesty drives resolution. By refusing to answer nearly three times as often as its base model, it paradoxically solved more problems because users trusted the validated responses it did provide. * Query resolution rose by 7.1 percentage points across 3,297 sampled customer queries. * Operational costs dropped to $0.001 per query, making it 20-50x cheaper than using GPT-4.1. * Processing speeds were 3-5x faster than frontier models, running efficiently on a single GPU. Ff news take: This announcement moves the needle by debunking the "bigger is better" myth in fintech AI. While the industry has been obsessed with massive parameter counts, Backbase proves that domain-specific calibration and calibrated refusal are the real keys to ROI. Achieving a 7.1% resolution lift while slashing costs by 50x is a massive win that finally addresses why 95% of AI pilots fail to impact the bottom line. Featured speakers.

Backbase acquires Kasisto, leading the industry shift to Agentic Banking. Strategic acquisition brings banking-grade agentic AI with governance and regulatory controls built into the Backbase AI-native Banking OS. July 2026 - Backbase has announced the acquisition of Kasisto, a pioneer in agentic AI for banking and financial services. Kasisto's agentic platform, financial services intelligence, and New York-based team are now part of Backbase and the AI-native Banking OS. Most banks have deployed agentic AI in isolated pockets - agents that answer questions without resolving work, leaving intent fragmented across channels, contact centers, and operations. Closing that gap requires purpose-built banking intelligence, reasoning-native agents, and governance embedded from the ground up - not generic AI platforms. Together, Backbase and Kasisto deliver the only AI-native solution built for the full complexity of agentic banking in regulated financial services. The acquisition is a defining step in Backbase's mission to deliver the Unified Frontline - one operating model where customers, employees, and AI agents operate as one with shared context, governed authority, and the same source of customer truth. With Kasisto embedded at the core of the AI-native Banking OS, banks can finally resolve customer intent through governed, intelligent execution. The strategic acquisition advances Backbase's three strategic priorities: * Banking-grade agentic AI, embedded in the Banking OS. Kasisto's agentic platform is built on the latest generation of reasoning-native AI - intelligent agents that understand context, apply judgment, and act within banking-specific governance and compliance controls. Combined with the Banking OS, banks can deploy agents that natively handle the full arc from customer intent to governed resolution across all conversational banking surfaces such as chat, messaging, and voice. * AI-native Customer Operations and front-to-back resolution. Many customer needs still break down between digital self-service, contact centers, servicing teams, and back-office operations. Kasisto's conversational and agentic AI capabilities, combined with the Banking OS, turn customer intent into governed execution - collecting evidence, checking eligibility, applying policy, triggering workflows, and resolving the need end to end. Banks move from agents that answer questions to agents that resolve work - including proactive, compliant outbound engagement before customer needs become inbound service demand. * Deepen US footprint. The acquisition unites complementary agentic AI capabilities, established customer bases across North America, and expert teams under one roof - strengthening Backbase's position in its largest strategic growth market. Jouk Pleiter, CEO and Founder at Backbase, said, "This acquisition sharpens our position as the strategic partner for banks serious about AI transformation. Kasisto brings proven agentic AI and deep financial services intelligence - moving us decisively into the era where customers express intent naturally and the bank resolves it through governed, intelligent execution. With Kasisto inside the Banking OS, no one is better positioned to lead the shift from conversation to resolution." Lance Berks, CEO at Kasisto, said,"Agentic AI will reshape banking over the next decade. Together, Backbase and Kasisto represent the convergence of the AI-native banking OS and purpose-built agentic AI and financial services intelligence, setting a new standard for how banks and financial institutions compete and win." The combined Agentic Banking suite is available immediately through the Banking OS to all current and future customers. About Backbase Backbase built the AI-native Banking OS - the operating system that turns fragmented banking operations into a Unified Frontline. Customers, employees, and AI agents work as one across digital channels, front-office, and operations. 120+ leading banks run on Backbase across Retail, SMB & Commercial, Private Banking, and Wealth Management. Recognized by Forrester, Gartner, and Datos as a category leader, Backbase was founded in 2003 by Jouk Pleiter and is headquartered in Amsterdam, with teams across North America, Europe, the Middle East, Africa, Asia-Pacific, and Latin America.