Amazon Web Services (AWS) released “Reimagine Turning AI into Value,” a report based on qualitative interviews with 150 C-level executives, to examine how organizations can deliver business value from AI projects. Its central message is that business outcomes should guide the work, which depends on organization design and change management as much as technology. The report organizes interview findings by topic and shares project examples and lessons from executives.

AWS identifies efficiency gains as a starting point, saying freed capacity should be redirected toward growth, preferably in the same quarter. It puts business outcomes ahead of adoption rates and output volume, recommends pairing efficiency measures with quality scores, and names judgment, problem framing and systems thinking as critical human skills. Other themes include proprietary data as a competitive advantage, placing governance controls in system architecture rather than human approval chains, and assigning AI costs to the unit deploying each project.

The report’s recommendations include tracking time, support cases and code alongside quality measures, since faster or higher-volume work does not automatically create value. It describes Amazon’s adoption-to-value stages as access and exploration, discrete tasks, iteration and context, reusable workflows, and process redesign. For hiring, it lists unlearning agility, problem framing, judgment over knowledge, systems thinking and curiosity.

For agentic AI governance, AWS recommends establishing identity, access and encryption basics, setting security boundaries outside the agent, beginning with human approval, allowing autonomy once systems are reliable, and conducting continuous testing. It also advises using FinOps to make costs transparent, assigning a unit responsibility for them, and holding people who define AI use cases accountable. Work redesign should specify intent, constraints and business decision rules, make recurring work repeatable in smaller units, leave intent and judgment to people, build verifiable processes and treat each redesign as a hypothesis. The report is AWS-focused, with many case studies drawn from Amazon programs, and highlights structural challenges without proven solutions, including the junior talent gap and governance of fleets of AI agents.

Amazon Web Services (AWS) released “Reimagine Turning AI into Value,” a report based on qualitative interviews with 150 C-level executives, to examine how organizations can deliver business value from AI projects. Its central message is that business outcomes should guide the work, which depends on organization design and change management as much as technology. The report organizes interview findings by topic and shares project examples and lessons from executives.

AWS identifies efficiency gains as a starting point, saying freed capacity should be redirected toward growth, preferably in the same quarter. It puts business outcomes ahead of adoption rates and output volume, recommends pairing efficiency measures with quality scores, and names judgment, problem framing and systems thinking as critical human skills. Other themes include proprietary data as a competitive advantage, placing governance controls in system architecture rather than human approval chains, and assigning AI costs to the unit deploying each project.

The report’s recommendations include tracking time, support cases and code alongside quality measures, since faster or higher-volume work does not automatically create value. It describes Amazon’s adoption-to-value stages as access and exploration, discrete tasks, iteration and context, reusable workflows, and process redesign. For hiring, it lists unlearning agility, problem framing, judgment over knowledge, systems thinking and curiosity.

For agentic AI governance, AWS recommends establishing identity, access and encryption basics, setting security boundaries outside the agent, beginning with human approval, allowing autonomy once systems are reliable, and conducting continuous testing. It also advises using FinOps to make costs transparent, assigning a unit responsibility for them, and holding people who define AI use cases accountable. Work redesign should specify intent, constraints and business decision rules, make recurring work repeatable in smaller units, leave intent and judgment to people, build verifiable processes and treat each redesign as a hypothesis. The report is AWS-focused, with many case studies drawn from Amazon programs, and highlights structural challenges without proven solutions, including the junior talent gap and governance of fleets of AI agents.