AI becomes a table-stake
SAP's Chief Quantum Officer explains that artificial intelligence is rapidly moving from a differentiator to a baseline capability. Within a few years, most large enterprises will have access to similar predictive tools, meaning that forecasting alone will no longer set a company apart.
The decision-making gap
The real challenge, he says, is not knowing what might happen but deciding what the business should do about it. In the final weeks of a financial quarter, accounts payable may hold payments to protect liquidity, accounts receivable may accelerate collections, and sales may pull forward deals or offer concessions. Each function makes a rational local choice, yet the combined outcome can be sub-optimal for the enterprise as a whole.
AI can flag likely wins, identify risky receivables and estimate the impact of a concession, but it does not answer the ultimate question: what should the company actually do? A discount may protect revenue while eroding margin; a quick dispute resolution may safeguard cash but signal weakness. These trade-offs ripple across finance, cash flow, delivery, risk and future customer value.
Enterprise Decision Computing emerges
To close the gap, a new technology category is forming: Enterprise Decision Computing. This approach turns a business decision, its possible actions, objectives, constraints, uncertainty and interdependencies, into a computable object that can be solved and optimised as a whole.
Traditional Enterprise Resource Planning systems execute processes, business intelligence explains the past and AI predicts outcomes. None of these, alone or combined, prescribe the coordinated set of actions a firm should take given its full set of goals and limits. Enterprise Decision Computing adds an enterprise-wide decision layer that can be continuously represented, governed, measured and improved.
Where quantum fits
The officer notes that most of the decision layers, revenue, closing probability, margin, payment terms, cash-flow timing, dispute status and delivery constraints, can be handled with classical optimisation and simulation today. However, as models become more interconnected and richly constrained, they can outstrip the capacity of classical methods without resorting to simplifying assumptions that dilute the answer.
At that point, quantum computing could provide the extra computational power needed to evaluate richer models without stripping away critical interactions. The technology is not a wholesale replacement for classical tools; it is a means to extend the depth of decision models once they reach a complexity that classical approaches cannot manage efficiently.
Steps for CEOs now
Decision debt, the officer warns, compounds like financial debt, quietly until it becomes a crisis. He advises senior leaders to identify a high-frequency, high-stakes domain where finance, sales or treasury currently operate in silos, run a baseline decision model and compare the coordinated outcome with the siloed result. The required investment is modest, but the cost of falling behind competitors who adopt such decision architecture could be significant.
The next competitive frontier, therefore, is not simply more data or more powerful AI, but the ability to build a decision architecture that captures the full complexity of an operating business and uncovers coordinated actions no single function could identify alone. Companies that understand where their current simplified decisions leave value on the table will be best placed to reap the benefits when quantum-ready solutions become commercially viable.