Executives Must Define AI Decision Rights Before Granting Autonomy

September 2, 2026 at 03:35 AM2 min

This digest was compiled by AI from multiple sources — links to the originals are below.

McKinsey's 2026 survey found 80% of respondents reported AI improved individual productivity, yet only 37% attributed EBIT impact to AI. The gap signals that local productivity gains do not automatically translate into enterprise value. Executives face the harder question of what decision rights to transfer as AI moves from assisting to executing.

Key Facts

McKinsey's 2026 global survey found 80% of respondents said AI improved their individual productivity.
Only 37% of respondents attributed at least some EBIT impact to AI, and about 6% qualified as AI high performers under McKinsey's definition.
The article identifies four distinct problems that can cause AI productivity gains to fail at the enterprise level: preference, representation, flow, and feedback.
As AI moves from assisting to recommending, deciding, and executing, companies transfer decision rights, not just work.

The Productivity Gap

McKinsey's 2026 global survey found that 80% of respondents said AI had improved their individual productivity. Yet only 37% attributed at least some EBIT impact to AI, and only about 6% qualified as AI high performers under McKinsey's definition. These findings reveal that local productivity and enterprise value are different outcomes. An AI system can make a task dramatically faster without improving the performance of the larger business system.

Four Underlying Problems

A preference problem occurs when decision-makers have not resolved what the organization wants when important priorities conflict. A representation problem arises when the trade-off has been resolved but did not carry through in the goals, constraints, metrics, uncertainty, or escalation rules available to the AI. A flow problem happens when the AI improves a part of the workflow that does not constrain overall performance. A feedback problem emerges when individual decisions seem reasonable but their effects accumulate or interact in ways the organization cannot see or correct quickly enough.

Decision Rights Transfer

As AI moves from assisting people to recommending actions, making decisions, and executing them, companies are transferring more than work. They are transferring decision rights, a shift that recent organizational research examines explicitly. The central question for executives is what they are giving the AI permission to decide.