Is The Window For AI Catch-Up Closing?

This article originally appeared on Forbes Business Council.

Artificial intelligence (AI) seems to be the only thing we discuss anymore. Its presence can be found woven throughout every news story, circling the workplace watercooler and sparking lively dinner-party debates. Heck, it may now rank higher than the weather when it comes to small talk. Yet, despite AI’s demand to occupy so much of our daily lives, business leaders around the globe continue struggling with its adoption.

A recent McKinsey report found that nearly two-thirds of companies in 2025 were just at the AI piloting stage. And only about 6% of respondents were what the consulting firm calls “high performers” (companies with integrated AI driving more than 5% of earnings). Moreover, recent research from the Boston Consulting Group (BCG) found that “only 25% of frontline employees said that they receive sufficient guidance from leadership on how to use AI effectively.”

In another survey from BCG, the firm polled 1,400 C-suite executives globally and found that 62% cited a shortage of talent and AI skills as their biggest challenge to achieving AI value (ahead of issues such as unclear priorities or lack of strategy). Yet only 6% said they have begun upskilling their workforce in a meaningful way.​

Those numbers are likely higher for many small and mid-sized enterprises, particularly in industries with complex systems where AI adoption must be closely measured to avoid major headaches like AI sprawl.

While it may sound counterintuitive, restraint in early AI adoption may have been the right decision for many companies, as AI has overcome significant barriers in the last several years. Early on, many tools promised dramatic gains, but organizations struggled to identify use cases that delivered measurable value in real operational environments.

Early AI adoption was often defined by experimentation, unclear returns, lack of governance and expensive pilot projects that never scaled. Many companies invested heavily without seeing a meaningful impact. Even worse, some faced serious consequences for projects that went horribly awry, like when tech firm Replit’s AI coding assistant acted out of line and wiped out the production database of startup SaaStr. Or when the Chicago Sun-Times and the Philadelphia Inquirer unwittingly featured summer reading list recommendations for books that didn’t exist. These mishaps only scratch the surface of failed AI initiatives.

Not only can business leaders learn from these costly and sometimes irreparable mistakes, but they can use such examples as compelling cases to build stronger governance, align AI initiatives with real business value and adopt more mature, proven technologies that reduce risk while increasing long-term competitive advantage.

Fast-forward to today, and AI use cases that deliver clear operational value are now widely recognized. Across small and mid-sized enterprises, applications such as predictive maintenance of equipment, automated quality checks, demand forecasting and optimization of scheduling or resource allocation have shown measurable returns. For example, IBM insights note that moving from preventive maintenance to a predictive model can lead to a 25% to 30% reduction in maintenance costs by performing repairs only when the data shows signs of actual wear.

Additionally, barriers to implementation are steadily falling as AI becomes more accessible for small and mid-sized companies. Cloud platforms, pre-trained models and AI capabilities embedded directly into software have helped reduce the technical complexity of adoption. Another BCG study of a global industrial goods company found that GenAI tools embedded directly into sourcing workflows enabled “50% faster document and revision drafting, faster and more consistent offer comparisons with improved risk detection, and 50% to 75% faster knowledge retrieval.”

What could this mean for organizations that operate with smaller, more focused teams? Access to the same kind of data that would have once required a robust crew of specialized data scientists, which leads me to my next point.

Over the next several years, I expect AI will increasingly become integrated into core operational systems. This could look like decision-making becoming more data-driven and, in some cases, partially automated. Planning processes will likely rely more heavily on predictive models. Equipment reliability may be monitored continuously through intelligent analytics.

Once such capabilities become standard across an industry, the competitive impact tends to become structural. As a recent Forbes article notes, “The introduction of AI into core business operations does more than speed things up. It changes the math of business itself.”​

So, where should small and mid-sized enterprise leaders start? First, remember that the objective is not to become an “AI company.” It is to improve the effectiveness of your existing operations.​ Your AI adoption strategy should be focused on value and not just flaunting the latest tools and trends. Remember: You’re not playing catch up, you’re creating measurable impact.

Here is a framework you can use to get started:

Above all, remember that adopting AI isn’t just about driving efficiency; it’s about creating long-term growth. If you took your time choosing the right path, that was a smart move. That patience gives you the advantage of learning from others and implementing the AI tools and strategies that best align with your business goals.​