MIT study: GenAI is big business but only 5% of businesses are getting a return on investment

Only 5% of organizations are receiving a return on their investment in AI, despite spending $30 billion to $40 billion on GenAI enterprise investment, a new MIT study says. 

According to the study, 95% of organizations from buyers, such as enterprises, mid-market and small businesses, along with builders, including start-ups, consultants and venders, are getting zero return on their investment. 

“This divide does not seem to be driven by model quality or regulation, but seems to be determined by approach,” the study states. “Tools like ChatGPT and Copilot are widely adopted. Over 80% of organizations have explored or piloted them, and nearly 40% report deployment. But these tools primarily enhance individual productivity, not P&L [profit and loss] performance.” 

Enterprise-grade systems, customer or vendor-sold tools are getting attention, with 60% of organizations evaluating the tools. However, only 20% of the tools reached pilot stage, and only 5% reached production. 

Brittle workflows, lack of contextual learning, and misalignment with day-to-day operations are the main reasons the tools fail, the report says. 

Through interviews, surveys, and analysis of 300 public implementations, four patterns emerged to define a term the report coins as the “GenAI Divide:” 

• • Only two of the eight major sectors have seen meaningful structural change;
  • Big firms lead in pilot volume but lag in scale-up;
  • Budgets favor visible, top-line functions over high-return-on-investment; and
  • External partnerships see twice the success rate of internal builds.

“The core barrier to scaling is not infrastructure, regulation, or talent,” the report says. “It is learning. Most GenAI systems do not retain feedback, adapt to context, or improve over time.” 

The small group of buyers who succeed requires process-specific customization and evaluates tools based on business outcomes rather than software benchmarks. 

“They expect systems that integrate with existing processes and improve over time,” the report says. “Vendors meeting these expectations are securing multi-million-dollar deployments within months.” 

According to the report, the tools have not driven headcount reductions. Yet, those who have crossed the GenAI Divide are seeing selective workforce impacts in customer support, software engineering, and administrative functions. 

Reduced spending for business process outsourcing is where businesses have seen the most measurable savings, the report says. Improved customer retention and sales conversions through automated outreach and intelligent follow-up are other areas businesses have seen benefits. 

Only two industries reviewed are seeing a disruption of reality on the right side of the divide, the report says. This is tech and media, while other industries remain on the wrong side of transformation. Other industries included energy and materials, advanced industries, financial services, consumer and retail, healthcare and pharmacy, and professional services. 

ChatGPT often beats enterprise tools because it is better, faster, and more familiar, even if similar models are used for both, the report says. 

A corporate lawyer told researchers that her company invested in a $50,000 tool specialized in contract analysis. Yet, she defaults to ChatGPT, saying that the “fundamental quality difference is noticeable.” 

“ChatGPT consistently produces better outputs, even though our vendor claims to use the same underlying technology,” the lawyer says in the report. 

However, ChatGPT is also limited, the report says. It forgets context, doesn’t learn, and can’t evolve. 

“For mission-critical work, 90% of users prefer humans,” the report says. “The gap is structural. GenAI lacks memory and adaptability.” 

According to the report, the best organizations and vendors succeeding are those solving for learning, memory, and workflow adaptation. 

About 50% of AI budgets are for sales and marketing, the report says. Yet, most of the dramatic cost savings come from back-of-office automation. 

Front-office tools tend to get the attention, but the real savings come from back-office tools, the report says. 

Front office AI tools are 40% faster with lead qualification speed and see a 10% improvement in customer retention. Yet, back-of-office tools see $2 million to $10 million in customer service and document processing, and agency spend reduction of about 30% in external creative and content costs. Risk checks for financial services save $1 million annually on outsourced risk management. 

In conclusion, the report says that the organizations that successfully cross the GenAI Divide do three things differently: they buy rather than build, empower line managers rather than central labs, and select tools that integrate deeply while adapting over time. 

“For organizations currently trapped on the wrong side, the path forward is clear: stop investing in static tools that require constant prompting, start partnering with vendors who offer custom systems, and focus on workflow integration over flashy demos,” the report says. “The GenAI Divide is not permanent, but crossing it requires fundamentally different choices about technology, partnerships, and organizational design.” 

IMAGE

Photo courtesy of jittawit.21/iStock