Futureproofing Your Business – People, AI and Productivity

AI's Uneven Productivity Surge

As 2026 begins, artificial intelligence has moved beyond hype into uneven real-world deployment. Companies that have integrated AI deeply report tangible benefits: global surveys indicate average net productivity rises of 11.5% and modest workforce reductions of around 4% among mature users. These gains stem from automating routine tasks and augmenting decision-making across functions.

Yet the broader picture reveals a stark divide. Despite trillions in cumulative investment and widespread experimentation—nearly 90% of organisations now use AI in at least one area—most executives report little to no measurable impact on productivity or headcount over recent years. Over 80% of firms in multi-country surveys see zero effect, reviving discussions of a modern productivity paradox reminiscent of earlier technology waves.

This lag matters because competitive pressure is intensifying. Early adopters pull ahead with efficiency advantages that compound over time, while laggards risk falling behind in cost structures and innovation speed. McKinsey and others project generative AI could contribute trillions annually to global output, but only if organisations move beyond pilots to redesign workflows—a transition many have yet to master.

Regulatory timelines add urgency. The EU AI Act's obligations for high-risk systems activate in phases through August 2026, demanding transparency, risk assessments and governance that many businesses are scrambling to implement. In the US, state-level rules on automated decision-making and frontier models take effect in 2026, even as federal efforts seek to streamline or preempt fragmented approaches. Non-compliance carries fines and operational restrictions.

Non-obvious tensions complicate the picture. AI often intensifies rather than reduces workloads by enabling faster paces and broader task scopes, sometimes extending hours without formal mandates. Entry-level roles face disproportionate erosion as AI handles foundational tasks that once built skills, raising concerns about long-term workforce development even as aggregate projections suggest more jobs created than lost—if reskilling occurs at scale.

Stakeholders diverge sharply: executives anticipate productivity boosts of 1-2% in coming years, while employees and some studies predict minimal or even negative employment effects without deliberate redesign of human-AI collaboration. The stakes include not just immediate financial performance but sustained organisational capability in an economy where AI-driven productivity increasingly separates winners from the rest.