The Washington Post’s WP Creative Group published an Investment Reports feature examining why, three years after generative AI became mainstream, so few companies are seeing meaningful returns. Drawing on more than 50 interviews with global business leaders, the report identifies the characteristics that separate successful AI implementations from expensive experiments. DataBank CEO Raul Martynek is among the executives featured.

The report opens with a sobering finding: an MIT study found that 95 percent of enterprise AI initiatives had yet to deliver ROI, while a Deloitte study found only 13 percent of even the most successful implementations saw returns within 12 months. The pattern emerging across industries is consistent — organizations that invest in data foundations, disciplined governance, and workforce preparation before layering in AI are outperforming those that rushed to adopt the technology without structural preparation.

“Beyond software development, we’re beginning to see compelling use cases in areas like healthcare and scientific discovery. However, businesses are still operating complex organizations with existing responsibilities and systems. Integrating AI into those environments will take time.”
— Raul Martynek, CEO, DataBank

Martynek also addresses the infrastructure dimension of AI’s future, noting that semiconductor innovations could dramatically improve efficiency in the years ahead, which will in turn influence how much data center capacity is actually needed. The report reinforces DataBank’s position at the center of the AI infrastructure story, framing the physical layer of compute, power, and connectivity as the foundation on which enterprise AI outcomes ultimately depend.

Read the full report at The Washington Post. Note: this content is paid for and supplied by Investment Reports. The Washington Post newsroom was not involved in its creation.