The companies that are most exposed to AI show the biggest gains in labour productivity and other metrics.

AI is becoming a fulcrum for company performance. PwC's latest AI Jobs Barometer report shows that the companies that are most exposed to AI show much stronger labour productivity than the least-exposed companies. What's more, the productivity gains are even more concentrated among a small set of superstar performers.

Labour productivity, measured by revenue per employee, is up across the board, but the increase is bigger for some companies than others. The difference?  A clear correlation with AI exposure. Companies operating in the most AI-exposed sectors (like software development, finance, and engineering) recorded 34% productivity growth in 2025 relative to a baseline of 2018. Meanwhile, the least-exposed companies increased productivity by 24%. That difference has been present every year since 2022, and the gap is widening.

Looking exclusively at the first group, a set of superstar top performers show even bigger gains. The top quintile within the most-exposed companies increased their labour productivity by 163% compared with 2018, nearly five times higher than the most AI-exposed companies overall.

These results align with PwC's recent AI performance study, which found that 20% of companies capture 74% of the gains from AI. What are the superstars getting right? In the performance study, the main factor is that top companies use AI to pursue growth opportunities, especially those created by collaborating or competing across sectors.

The results have clear implications for business leaders:

Focus on value-creation over automation. Companies gaining the greatest value from AI aren’t using it only to cut costs or shave headcount. Instead, they’re using it to unlock new revenue, enter new markets, and create new forms of value. Leaders should frame their AI agenda around growth opportunities, especially by partnering across industries.

Build AI foundations. Our AI performance study shows that having the right AI foundations—for example, a targeted strategy to scale the most valuable AI initiatives and modernised data and technology platforms to support them—greatly increases the ability to capitalise on AI’s promise.

Invest in agentic AI. AI agents help workers use their uniquely human expertise to deliver value at much greater scale. This enables organisations to think, adapt, and execute faster than competitors. Indeed, PwC’s AI performance study found that those companies getting the most value from their AI investments are twice as likely to deploy agents as those seeing lower ROI.