Most companies have access to the same AI tools. Most companies are not getting the same results. According to PwC, that gap is now measurable, and it’s large.
According to PwC, 20% of organizations capture roughly 74% of the technology’s economic value. PwC found that organizations it classifies as AI leaders are 2.6 times more likely to deploy AI toward business model reinvention rather than cost reduction alone. These figures come from PwC’s own research and carry that attribution throughout, they represent PwC’s methodology and definitions, not an independent industry standard.
The distinction PwC draws between reinvention and cost reduction is worth unpacking. Cost reduction is what most AI deployments look like in practice: automating a support queue, speeding up document review, replacing a manual data entry process. These are real gains. They show up in efficiency metrics. But they don’t change what a company does or who it serves. Reinvention is different. It means using AI to build products or revenue streams that weren’t possible before, or to enter markets that were previously inaccessible at the company’s scale.
The 2.6x figure tells you that the gap between these two approaches isn’t a marginal difference in ambition. It’s a structural divergence in how leadership teams are thinking about what AI is for.
The 80% of companies not capturing proportional AI value aren’t failing because they lack access to tools. They’re failing because they’re using tools to do the same things faster rather than to do different things entirely. That’s a strategic choice, often made under near-term pressure to demonstrate ROI. It’s rational at the individual decision level. It compounds into a significant competitive disadvantage at the portfolio level.
For investors and analysts, the PwC framing offers a useful diagnostic. AI maturity isn’t a binary, it isn’t whether a company uses AI or not. It’s whether AI is touching the revenue model or staying inside the cost structure. Companies in the second category may look efficient in the short term and underprepared for disruption in the medium term.
The market signal here is directional rather than definitive. PwC’s “leader” classification is their own framework, not an independent benchmark, and the specific figures (74%, 2.6x) require the full report for methodology context. Watch for the full PwC publication for the underlying survey population, sample size, and how “economic value” is operationalized. Those details determine how broadly these findings can be applied.
What’s not in doubt is the direction. AI value is concentrating. The question for any executive or investor reviewing their AI portfolio is whether their organization is in the 20% or the 80%, and if the latter, why.