A recent MIT Media Lab study titled “The GenAI Divide: State of AI in Business 2025” dissects enterprise AI adoption and finds a startling truth: 95% of businesses see no measurable return on their AI investments. Despite pouring in an estimated $30–$44 billion, most pilot projects never generated tangible profit or productivity gains.

The study analyzed around 300 public AI deployments, engaged 150 executives, and surveyed 350 employees, concluding that most AI efforts remain stuck in the experimental phase. Common pitfalls include flawed integration into workflows, misalignment with business operations, and a lack of clear, outcome-driven planning.

Yet, the report also highlights a successful minority: the 5% of AI pilots that delivered value. These initiatives thrive by focusing on one specific business problem, partnering with specialized vendors, and improving back-office operations such as document handling and risk monitoring.

MIT warns that this growing “GenAI Divide” isn’t technology-driven - it’s a failure in strategic execution. Organizations succeeding with AI aren’t just chasing hype—they’re aligning tools with measurable business outcomes.