The AI productivity paradox

Executive summary

Most organizations have deployed AI tools. Most are seeing modest returns. The gap between those capturing outsized value and those that aren't isn't a technology problem - it's a leadership problem. This week: the evidence, the framework, and what leaders must do differently.

Top leadership insight

The 3% vs. 55% gap

New research found that 80% of software engineers using AI tools see productivity gains of roughly 3%. The top 20% see gains of 55% or more. Closing that gap across the global engineering workforce represents approximately $800 billion in untapped value. The disparity isn't about access to tools. It's about the conditions leaders create for people to use those tools well: skill development, accountability, and clarity of purpose.

AI and leadership update

The age of intention

At a major executive development conference this month, a central theme emerged: AI is entering the "age of intention." The technology can sharpen thinking, test assumptions, and surface blind spots - but it cannot exercise judgment. Leaders who use AI as a thinking partner, not a thinking replacement, see dramatically different results. The question isn't whether your organization has AI. It's whether your leaders are intentional about how it's used.

Leadership tip

Before your next AI initiative, run a leadership pre-mortem. Ask: "If this fails in six months, what are the most likely reasons?" Design your accountability structures around those answers.

Framework of the week

Double-loop learning

From the work of Chris Argyris

Single-loop learning corrects errors within existing rules. Double-loop learning questions the rules themselves.

Most AI adoption is single-loop: deploying tools to optimize what already exists. Leaders ask "are we doing this right?" but rarely "should we be doing this at all, and how do we measure real value?"

The organizations seeing 55% productivity gains aren't just using AI better. They're questioning what productivity means, who owns outcomes, and what beliefs about work they may be unconsciously protecting. That shift, from efficiency to intention, is where AI adoption becomes transformation.

Statistic of the week

3% vs. 55% - Productivity gap: average vs. top AI tool users.

The top 20% of AI tool users in engineering capture 18 times more productivity gain than the other 80%. The gap is a leadership and capability issue, not a technology issue.

Source: McKinsey Technology Trends Outlook, October 2026

Middle East and Saudi lens

Saudi Arabia: ambition ahead of execution

A new AI maturity report finds that 60% of Saudi organizations report large productivity gains from AI, versus 46% globally. Yet the report identifies a clear gap: execution lags ambition. Saudi leaders score high on AI vision and accountability. The next frontier is tighter execution discipline, measurable ROI, and scaling successful pilots beyond the proof-of-concept phase.

Closing insight

The productivity paradox has a straightforward diagnosis: we measure AI adoption, not AI outcomes. When leaders start asking "what changed because of AI?" instead of "how many people are using it?", the results shift. Accountability, not technology, is the real unlock.

Zahi Abdein