The GenAI Divide: Why Most Companies Aren’t Seeing Value from Generative AI

Despite the hype surrounding generative AI, a recent MIT study reveals a significant gap between investment and realized value. The study, titled “The GenAI Divide: State of AI in Business 2025,” found that a staggering 95% of companies are not extracting meaningful value from their generative AI initiatives.

Key Findings from the MIT Study

• Limited Success Rate: Only 5% of generative AI pilots deliver measurable business value or rapid revenue growth. The remaining 95% either stall or fail to demonstrate a business impact.
• Organizational Learning Gap: The primary obstacle isn’t technical limitations, but rather an organizational “learning gap.” Companies struggle to integrate AI into existing operational workflows and adapt processes to effectively leverage these new tools.
• Strategic Partnerships & Niche Solutions: The highest success rates are achieved by companies that focus on purchasing niche, specialized AI solutions rather than attempting to build everything in-house. Forming strategic partnerships also proves to be a key differentiator.

The Research Behind the Findings

The MIT Media Lab and Project NANDA conducted the research, reviewing over 300 public AI implementations. The study also included interviews with 52 organization leaders and a survey of 153 executives from major industry events.

Implications for Businesses

The findings suggest that simply investing in generative AI technology is not enough. Companies must prioritize organizational change, process adaptation, and strategic partnerships to unlock the potential benefits of AI. A focused approach, solving specific business problems with specialized solutions, is more likely to yield positive results than broad, in-house development efforts.

As Davide Fabrizio noted in a LinkedIn post discussing the study, the divide highlights the need for companies to move beyond experimentation and focus on practical implementation to realize the promised value of generative AI.

Source: LinkedIn – Davide Fabrizio