PwC published its 2026 AI Performance Study in April. The firm surveyed 1,217 senior executives at large companies spanning 25 sectors. What they found should change how business leaders think about their AI spending: 74% of AI's economic value is being captured by just 20% of organizations. The other 80% are splitting the remaining 26% among themselves. The divide traces to one thing: how they work.
What PwC Measured
The study went beyond survey data on AI adoption rates. Researchers asked companies two things: how much AI-driven revenue and efficiency gain they were seeing, and exactly how they deployed the technology. That second question is what made the findings useful.
Most AI coverage measures adoption. This study measured performance. And the divide it found is not small.
Nearly three-quarters of AI's total economic value is flowing to one-fifth of organizations. The remaining 80% share what is left. At current rates of investment, Gartner research suggests why that disconnect persists: only one in five AI investments delivers any measurable return on investment. One in fifty delivers what Gartner calls transformational value.
High spending does not fix this. The money is going in. For most, it is not coming back out in proportion.
of AI's economic value is captured by just 20% of companies. The remaining 80% share 26% of the gains.
Source: PwC 2026 AI Performance Study, 1,217 executives across 25 sectorsWhat the Top 20% Do Differently
PwC did not find that leaders were using better AI tools. They were working differently.
The companies generating the strongest returns are twice as likely to redesign their workflows around AI rather than layering AI tools onto existing processes. That distinction carries more weight than any other finding in the study.
Think about what layering looks like. An employee gets access to a chatbot. They draft emails faster. Summarize meeting notes in seconds. The underlying workflow (how decisions get made, how information flows, how the work is organized) stays exactly the same. The tool is faster. The system is unchanged.
Workflow redesign looks different. A company asks: what decisions currently made by a person could be made more accurately by a system? What information flow creates delays that AI could remove? What approval process could be compressed or automated without increasing risk? The tool serves a new way of working, not a faster version of the old one.
The results speak. Companies in the top 20% generate 7.2 times more AI-driven revenue and efficiency gains than other respondents. They are also 2.8 times more likely to have increased the number of decisions made without human intervention (and they are doing this while maintaining or improving governance standards, not bypassing them).
The tool is faster. The system is unchanged. That is what 80% of AI investment looks like today.
The Tool Trap
Why do most companies stay in the 80%? The answer is that adoption is visible and proficiency is not.
Organizations buy AI subscriptions, roll them out across teams, measure adoption rates, and call it an AI strategy. These metrics look good in board presentations: active users, hours saved, licenses deployed.
What those metrics do not measure: whether the underlying way work happens has changed at all. A team using AI to speed up existing tasks is not the same as a team that has redesigned its work so AI handles entire decision loops. One is faster. The other compounds.
Gartner's finding that only one in five AI investments produces measurable ROI reflects this. The investment is real. The adoption is real. The proficiency needed to turn adoption into structural performance gain is missing.
The Proficiency Layer
The 7 Levels of AI Proficiency framework is built around exactly this distinction.
There are seven stages of AI capability, from basic prompt use at Level 1 through full system design and organizational coordination at Level 7. The divide PwC found between the top 20% and the rest maps directly onto the framework. Companies in the 80% tend to be operating at Levels 1 through 3: using AI for individual tasks, getting faster at existing work, but not changing the underlying structure of how work gets done.
Companies in the 20% are operating at Levels 4 through 6. They redesign processes around AI. They make system decisions, not just tool decisions. They build AI proficiency as an organizational capability, not a personal skill.
That distinction runs deeper than technology. It sits in how the organization thinks about what AI is for.
Related reading: Level 4: The Commander in the 7 Levels of AI Proficiency.
What This Means for Your Company
Indiana Governor Braun's IN AI initiative, designed to reach one million Hoosier employees, is built on the right instinct. Mid-market companies across the state need practical AI capability, not just awareness campaigns. But the PwC data makes something clear: the method is as consequential as the intention.
Training that teaches people to use AI tools is not the same as training that teaches organizations to redesign work around AI. One produces faster individual contributors. The other produces organizations that compound over time.
Companies in the top 20% measure AI success by what changed in how work actually happens: which meetings got eliminated, which approval cycles shortened, which decision loops now run without waiting for a human.
The PwC study was direct about the trajectory: organizations that begin workflow redesign now build a compounding advantage. Leaders learn faster, scale what works, and expand the number of decisions AI handles. Each redesigned process generates better data, which opens the next round of improvements.
The question worth asking right now: is your company adding AI to existing work, or changing the work itself?
Frequently Asked Questions
What is the PwC 2026 AI Performance Study?
The PwC 2026 AI Performance Study is a global survey of 1,217 senior executives at director level and above, spanning 25 sectors across multiple regions. Published in April 2026, it measured AI-driven revenue and efficiency gains and identified what distinguishes high-performing companies from the rest. The study found that 74% of AI's economic value flows to just 20% of organizations.
What do the top 20% of AI-performing companies do differently?
According to PwC's research, companies generating the strongest AI returns are twice as likely to redesign their workflows around AI rather than layering AI tools onto existing processes. They are 2.8 times more likely to increase the number of decisions made without human intervention. They generate 7.2 times more AI-driven revenue and efficiency gains than other respondents.
How can I find out if my company is in the top 20% or the 80%?
The free 7 Levels of AI Proficiency assessment at assess.launchready.ai places your team across seven capability stages, from basic prompt use through full system design and coordination. Companies in the top 20% typically operate at Levels 4 through 6, redesigning work around AI rather than using it for individual tasks within unchanged processes. The assessment takes under ten minutes.
Sources
- PwC. "PwC 2026 AI Performance Study." pwc.com
- Harvard Business Review. "9 Trends Shaping Work in 2026 and Beyond." February 2026. hbr.org
- Gartner. "Gartner Says CFOs Need to Rethink the ROI of AI Investments." March 24, 2026. gartner.com
- Indiana Capital Chronicle. "Braun Unveils Artificial Intelligence Business Portal." April 28, 2026. indianacapitalchronicle.com
- Indiana Governor's Office. "Governor Braun Announces IN AI to Grow Jobs and Wages Through Human-Centered AI." in.gov
Find your AI Proficiency level
The free 7 Levels assessment places you across seven stages of AI capability. Under ten minutes. Research-backed scoring.