If you run a business and you're wondering whether AI is real or just hype, this page is for you. I've pulled together what the research actually says — no jargon, no sales pitch — so you can make your own informed decision.
Here's the short version: AI is real, but most companies are wasting money on it. The difference between businesses pulling ahead and businesses spinning their wheels isn't budget or technical talent — it's strategy. The research below makes the case. The final section is how I think about helping clients get started without boiling the ocean.
The productivity gains from AI are well documented. Stanford and MIT economists tracked 5,179 customer support agents using a generative AI assistant and found a 14% productivity increase on average, with 34% gains for newer employees (Brynjolfsson, Li & Raymond, Quarterly Journal of Economics, 2025). The headline isn't the average — it's that the biggest gains went to people doing structured, repeatable work. That pattern matters when we get to where AI actually creates value.
But here's the disconnect. MIT's Project NANDA released its State of AI in Business 2025 report in July 2025. After studying organizations that have collectively spent $30–40 billion on AI, the researchers found 95% of organizations report no measurable P&L impact from their AI investments. McKinsey's State of AI 2025, surveying nearly 2,000 organizations across 105 countries, confirms the gap: 88% of organizations now use AI in some form, but only about 6% are pulling decisively ahead.
The gap isn't between companies that have AI and companies that don't. It's between businesses using it strategically and businesses just dabbling.
A RAND Corporation report from August 2024 found that more than 80% of AI projects fail — roughly twice the failure rate of traditional IT projects. RAND researchers interviewed 65 senior data scientists and engineers about why. The top cause wasn't technical. It was strategic misalignment — leadership having an unrealistic view of what AI can do, or no clear connection between AI projects and how the business actually runs.
The downstream cost shows up in employees' inboxes. Stanford's Social Media Lab and BetterUp Labs surveyed 1,150 U.S. workers in September 2025 and coined the term "workslop" — AI-generated work that looks polished but lacks substance. 40% of workers reported receiving it in the prior month, at a cleanup cost of roughly two hours and $186 per employee per month. For a 10,000-person company, that's about $9 million a year — the cost of using AI without a plan.
This tracks with what I see working with businesses directly. The companies that succeed with AI don't start with the technology. They start by understanding their own operations well enough to know where AI creates real value.
Most of the AI hype focuses on customer-facing applications — chatbots, personalized marketing, automated support. That's not where I recommend most clients start. The data backs this up: the MIT NANDA report found that more than half of corporate generative AI budgets are going to sales and marketing tools, while the highest-ROI deployments are in back-office automation — eliminating outsourced business processes, cutting agency costs, streamlining internal operations.
The Stanford-MIT customer service study makes the same point in a different way. The biggest productivity gains went to less-experienced workers handling routine, structured tasks — not the seasoned experts dealing with ambiguous customer issues. AI is most reliable on work that's repetitive, internal, and not exposed to customers.
There's also a buy-vs-build signal worth knowing. MIT NANDA found that companies that buy AI tools from specialized vendors succeed about 67% of the time, while companies trying to build in-house succeed at roughly a third of that rate. For an operations-heavy business that doesn't have an AI engineering team, that's a meaningful piece of guidance: focus on choosing the right tools and integrating them well, not on building from scratch.
The "fast follower" strategy has worked with previous technology cycles. With AI, the gap between leaders and laggards is widening faster than usual. BCG's Build for the Future 2025 report — based on a global survey of 1,250 senior executives across nine industries — found that AI leaders are growing revenue at twice the rate of laggards and capturing 40% more cost savings. McKinsey's data shows the same shape: a small group of about 6% of companies pulling away while the rest are stuck in pilots.
You don't need to be first. But the businesses that will come out ahead are the ones building a plan now, not the ones waiting for perfect conditions.
The simple version of everything above: AI works, most companies fail at it, and the difference is strategy — not budget, not tech.
When I work with a client, I don't try to AI-enable the whole business at once. We pick one or two operational areas where AI actually fits — usually somewhere in the back office, where the work is structured, internal, and measurable. We get a clean win there. That win funds the next one, and over a few moves we build a strategy for where AI fits across the rest of the business.
This approach is deliberate. It avoids the failure modes the research keeps surfacing — scope too big, no clear connection to the P&L, leadership chasing the wrong end of the business. And it builds the muscle to do AI well over time, instead of betting the year on one transformation project.
If that sounds like the right pace for your business, let's talk.
Working hands-on with clients across five Southeast markets.
Don't see your city? Join the waitlist →
That's exactly the conversation I have with every client. Let's figure out where AI fits in your business.
Let's Talk