Enterprise AI has entered a new phase where meaningful return on investment (ROI) is becoming the norm rather than the exception. AWS CEO Matt Garman revealed in a recent interview that an overwhelming majority of enterprise leaders now report positive financial outcomes from their AI initiatives, a stark contrast to the cautious experimentation seen just a year ago.
This accelerated value creation partly stems from enterprises leveraging cloud infrastructure already in place, allowing AI adoption to outpace even the long transition seen with cloud workloads alone. Garman emphasized Amazon’s substantial capital investments, which prioritize reliable, high-value assets like land, power, and servers. These commitments rely on careful forecast and customer demand visibility, minimizing financial risk and maximizing return on invested capital.
Beyond infrastructure, AWS supports enterprises in shifting from trial phases to production by guiding them to match AI models carefully to specific tasks—contrary to the costly practice of defaulting to the largest, most powerful models for every job. Tools such as Kiro, Amazon’s agentic development environment, optimize this process by routing tasks to appropriately sized models. This targeted approach notably reduces expenses while maintaining output quality.
Garman also advised companies to shift their focus from technical usage metrics to outcome-based evaluation. Encouraging employees to treat AI resources as investments rather than consumption limits helps drive smarter, value-focused decisions. The fastest route to meaningful ROI, he said, involves doubling down on successful AI projects and swiftly discontinuing those that fail to deliver results.