DBS Bank generated approximately SGD 1 billion — roughly $740 million — in economic value from its data analytics and artificial intelligence programs in 2025. Euromoney named it Asia's Best Digital Bank this month, citing the result as the clearest evidence that AI in financial services has moved from experiment to infrastructure.

The headline number is significant. What sits behind it is more instructive.

DBS has deployed more than 2,000 AI and machine learning models across the bank. Over 430 individual use cases are now in active production. These are not pilots. They are not proofs of concept awaiting budget approval. They are running systems handling real decisions across credit risk, wealth management, customer service, operations, and compliance.

The bank's approach was not to find one large AI application and announce the savings. It was to build a platform that could absorb AI across functions systematically, with governance structures that allowed rapid deployment without losing control of outcomes. That platform is now generating scale returns precisely because the infrastructure work was done first.

Most organizations are at the opposite end of this curve. They have AI tools in use across their business — some productive, some not — but no architecture that connects them, measures their impact, or allows them to be scaled. The AI is present. The return is not.

The DBS result points to something that has been said for two years but that few businesses have acted on: AI value compounds when the underlying data and governance architecture is built to support it. Without that foundation, you can run dozens of pilots and still report zero measurable outcome at the enterprise level.

For most businesses, the architectural question is not whether to invest in AI infrastructure. It is how to get useful work out of AI now, while building toward that infrastructure over time. That requires tools that connect to existing systems without requiring a multi-year re-platforming project first.

Viktor lives inside Slack and Microsoft Teams. You @mention it in a thread the same way you would ask a colleague. The output — a PDF, a report, a task created in your CRM, an email drafted in Gmail — lands where it should land.

The entry point is straightforward: identify the highest-frequency manual task in one function — report preparation, briefing notes, first-draft proposals, client research summaries — and automate that first. Measure the time saved. Use the result to make the case for broader deployment. The DBS model took a decade to produce a billion-dollar outcome. Your model does not have to start there. It has to start somewhere concrete and measurable.

The gap between organizations capturing AI value and those that are not is widening. DBS is the clearest current example of what the wide end of that gap looks like from the top.

For finance and professional services teams evaluating AI tooling, compliance is the first question. Viktor holds SOC 2 Type II certification and complies with GDPR, CCPA, and CASA Tier 3. Credentials are stored in an encrypted vault and are never used to train models. Sensitive actions can require Slack-based human approval before execution. Full details are at viktor.com/security.

You get $100 of free credits to begin. No time limit, no commitment. That's enough to do real work and see what Viktor can actually do before you spend a penny. There's also $50 off your first bill. You must use this exact link to receive both benefits: AIThatDelivers.com

Disclosure: Some links in this article are affiliate links. If you choose to get started with Viktor using the links provided, I may receive a commission — at no additional cost to you. I only recommend tools I use and believe in.