SAS experts outlined 13 banking predictions for 2026, from agentic AI in production to data integrity vaults.
Data and AI company, SAS, predicts that 2026 will mark a shift from banking AI experiments to higher-stakes deployments, where autonomous agents handle real customer requests, synthetic data introduces new integrity risks, and trust becomes a measurable performance metric.
In a set of 13 predictions, SAS experts described what they see as the breakthroughs and breaking points likely to shape banking, and the areas where institutions may be unprepared for the pace of change.
SAS’s 2026 outlook centres on a shift from model-driven to proof-driven intelligence, arguing banks will need to demonstrate transparency and evidence behind AI decisions as automated decisioning expands. It expects agentic AI to move into production, with governed, explainable systems handling customer requests, orchestrating workflows, and helping banks turn pilots into operational value.
It also warns banks will inherit new dispute and fraud pressures from agentic commerce, as AI agents make unauthorised purchases and criminals hijack or imitate legitimate agents. SAS forecasts banks will need to authenticate both humans and AI agents, while also addressing synthetic contamination risks by protecting golden-source data in controlled vaults and tightening governance over how GenAI interacts with critical datasets.
Across risk and markets, SAS expects an escalation in AI-enabled romance scams, a shake-up in financial crime technology toward cloud-native, AI-driven AML and fraud tools, and broader use of AI in quantitative credit and bond pricing, with bubble-aware modelling likely lagging adoption. It also anticipates regulated stablecoin pilots for cross-border settlement, retail banks scaling commerce media models, stronger climate risk stress testing, and the first meaningful production impacts of hybrid quantum-classical computing in risk and fraud
Stay updated on the crypto space by following our socials
