The State of AI in Business

(This is a follow-up post to The State of AI in Practice.)

The internet (or at least my neighborhood of it) is blowing up over the recent MIT study that finds: “Despite $30–40 billion in enterprise investment into GenAI, this report uncovers a surprising result in that 95% of organizations are getting zero return.”

The GenAI Divide: State of AI in Business 2025

That headline generates a lot of attention, but it takes a negative view of the data and misses many of the more important points from the research.

We shouldn’t be surprised, and we should actually be encouraged by the state and progress of AI already.

1. 25% of Major Sectors Show Meaningful Structural Change! And 5% of Integrated AI Pilots Are Already Extracting Millions in Value!

The paper says that “only 2 of 8 major sectors show meaningful structural change” and calls that “limited disruption”.

It also states that: “Just 5% of integrated AI pilots are extracting millions in value, while the vast majority remain stuck with no measurable P&L impact.”

Is that “only” and “just” as the paper states or is that: “Wow, 25% of major sectors show meaningful structural change already! And 5% of integrated AI pilots are already extracting millions of dollars in value!”

 We have too high expectations if we expect it to happen much faster. We’re so early in the adoption and development of this technology. Look at the difference every few months in the capability of the models and tools. And most companies don’t move that fast. These numbers are already proving that AI works. The numbers will only go up.

2. Individual Productivity Is Already Growing and Investments May Still Return P&L Value

“Despite $30–40 billion in enterprise investment into GenAI, this report uncovers a surprising result in that 95% of organizations are getting zero return.”

Apparently, zero return means that AI hasn’t affected “P&L performance” yet, but has affected “individual productivity”.

“Tools like ChatGPT and Copilot are widely adopted. Over 80 percent of organizations have explored or piloted them, and nearly 40 percent report deployment. But these tools primarily enhance individual productivity, not P&L performance.”

Individual productivity is a great place to start. If individuals are getting more productive, then that’s likely a leading indicator that at some point revenue will go up and expenses will go down.

And, like I always tell my kids, “It’s not that you’re not good at tennis. It’s that you’re not good at tennis yet.” Much of this investment hasn’t paid off yet, but still may pay off in the future, especially if we learn as we go.

3. Adoption Is Unstoppable

“While only 40% of companies say they purchased an official LLM subscription, workers from over 90% of the companies we surveyed reported regular use of personal AI tools for work tasks. In fact, almost every single person used an LLM in some form for their work.”

These numbers speak for themselves, especially when workers are finding a way to use it even where their companies don’t have an official solution.

4. Lessons to Take Away

Most importantly, the lesson isn’t that “95% of organizations are getting zero return”, but that AI can produce massive value, especially if you do it correctly:

• “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.”
• Companies “who succeed demand process-specific customization and evaluate tools based on business outcomes rather than software benchmarks. They expect systems that integrate with existing processes and improve over time.”
• “The highest-performing organizations report measurable savings. … Others cite improved customer retention and sales conversion through automated outreach and intelligent follow-up systems. These early results suggest that learning-capable systems, when targeted at specific processes, can deliver real value, even without major organizational restructuring.”

Yes, we can do it better. And yes we will. A lot of money invested will be wasted. A lot of it will also pay off over time. Let’s keep learning.