HOW DO WE UNDERSTAND WHERE AI CAN CREATE VALUE?
Start with business economics rather than the technology itself. Identify where AI may affect unit costs, throughput, revenue quality, decision speed or risk, then define a baseline and measurable outcome. Value is more credible when the use case addresses a material constraint, fits the operating context and accounts for implementation costs, governance requirements and adoption effort.
Each initiative should connect to a testable economic driver, separating theoretical potential from observed results.
HOW CAN AI CHANGE THE OPERATING MODEL?
AI can change an operating model when it is integrated into workflows rather than added as a standalone tool. It may redistribute tasks between people and systems, alter decision times and responsibilities, and require new skills or controls. The effect depends on process quality, data and governance, as well as the organization's ability to adopt the new design.
Automating a weak process can preserve its flaws; selective workflow redesign should precede scaling.
WHEN DOES AI CREATE A REAL COMPETITIVE ADVANTAGE?
Access to widely available models is rarely enough to differentiate a company. Advantage may emerge when AI improves a system of activities that is difficult to imitate: relevant, well-governed data, domain knowledge, distribution, customer relationships and learning speed. Its durability should be tested over time because competitors' tools, costs and practices continue to evolve.
Differentiation depends on execution and context; proprietary data and capabilities may contribute but are not automatically exclusive.
HOW CAN AI ERODE ENTERPRISE VALUE?
AI Value at Risk describes an exposure, not a certain outcome. It may rise when AI lowers entry costs, makes previously differentiated offers easier to compare, or reduces demand for some capabilities. The impact depends on the industry's adoption rate, customer responses and the company's ability to adapt its products, pricing and processes.
The analysis considers scenarios for revenue and margin pressure, their likelihood and the responses available to the company.
WHAT SHOULD CEOS AND BOARDS EVALUATE BEFORE INVESTING IN AI?
CEOs and boards should clarify the economic problem, expected outcome and accountable owner. The evaluation includes data quality and usage rights, process integration, security, compliance, vendor dependence and available skills. Defined metrics, total costs and review criteria help distinguish a useful experiment from an initiative that is ready to scale.
Oversight connects accountability, risk and committed capital, with review proportionate to the initiative's significance.
HOW DOES ENTERPRISE VALUATION CHANGE IN THE AI ERA?
AI adoption alone says little about enterprise value. The analysis considers whether use cases may affect cash flows, revenue durability, capital requirements and risk, including execution costs and timing. It also considers competitive exposure: the same technology may support productivity or reduce differentiation, depending on the sector, the company's position and how effectively it responds.
Operational evidence matters more than announcements: results, economic sustainability and execution capacity should be assessed together.
HOW SHOULD A COMPANY CHOOSE THE RIGHT AI TOOLS?
The choice should start not with the number of available features, but with the problem the company wants to solve. An AI tool is relevant when it fits existing processes, uses appropriate data, delivers a measurable benefit and has costs, risks and adoption requirements that suit the organisation.
The assessment should therefore begin with the use case — productivity, sales, customer service, finance, legal, operations or knowledge management — before comparing available solutions. There is no single best tool: the right choice is the one most closely aligned with the company's process, context and economic objective.