AI is the primary leadership topic of this moment. Over the past 30 years, I have had the opportunity to serve thousands of leaders across enterprise, non-profit, ministry, and government — and I challenge leaders to recognize that one of their primary responsibilities is to set the tone in every season and stage of innovation. This is a defining leadership moment — one that offers real opportunity, and for many leaders, real intimidation. As you evaluate how, when, where, and why to leverage the capabilities AI offers, here is some practical guidance on the current state across enterprise, and a roadmap for building an AI strategy that succeeds.

The Current Landscape

Enterprises are pouring money into AI at a pace that has no real precedent. Global AI spending is projected to hit $2.59 trillion in 2026, up 47% from $1.76 trillion the year before. Inside individual companies, AI is now consuming up to a third of innovation budgets.

Yet today, almost none of that spending is translating into profit. McKinsey's most recent global AI survey found that 88% of organizations now use AI in at least one business function, but only 39% can attribute any EBIT impact to it at all, and most of those say it accounts for less than 5% of their bottom line. Just 6% of companies qualify as genuine "AI high performers" seeing meaningful financial return. MIT's research on generative AI pilots found that 95% fail to deliver measurable ROI. Gartner projected that 30% of generative AI projects get abandoned entirely after the proof-of-concept stage.

This is the adoption-investment gap, and it is the single most important thing to understand before building an AI strategy: the money is moving faster than the value. Universal adoption and negligible impact are happening at the same company, often at the same time. That is not a technology problem. Every study cited above lands on the same root cause — organizational, not technical. Pilots get built and never integrated into real workflows. Ownership is unclear. Tools get bought before processes get redesigned. Leadership treats AI as a procurement decision instead of an operating-model change.

Building An AI Strategy

If that is the actual cause, then the fix has to start upstream of any tool selection. Here is what a real AI strategy needs to account for.

Start with the workflow, not the model. 

Most failed pilots aren't failing because the underlying AI is weak — they're failing because they were bolted onto an existing process instead of replacing or redesigning it. Before selecting a vendor or a use case, map the actual workflow the AI is meant to change, including who owns each step today and what breaks if that step changes. If you can't answer who is accountable for the outcome once the pilot ends, you don't have a strategy yet — you have a pilot.

Pick fewer bets and fund them properly. 

The "spray and pray" pattern — a dozen scattered pilots, none with a real budget or a real owner — is exactly the pattern MIT and Gartner describe as the highest-failure-rate approach. The 6% of companies McKinsey identifies as high performers didn't run more pilots; they ran fewer, better-resourced initiatives tied to specific P&L outcomes, with executive ownership that outlasted the pilot phase.

Measure EBIT impact, not activity. 

Usage metrics, adoption rates, and "number of employees using AI tools" are the wrong scoreboard. They're the metrics behind the 88% adoption figure that coexists with a 39% impact figure. A real strategy defines, before launch, what dollar or percentage outcome the initiative is meant to move, and tracks that number, not engagement.

Budget for integration, not just infrastructure. 

Most enterprise AI spending today follows the hyperscaler pattern — infrastructure and tooling first. But the studies above are consistent that integration, change management, and workflow redesign are where the value actually gets unlocked or lost. If your AI budget is 90% tooling and 10% implementation, you're funding the part of the problem that's already solved and starving the part that determines whether you see any return.

Give initiatives a life beyond the pilot. 

Thirty percent of generative AI projects get abandoned after proof-of-concept, often because no one was responsible for what happens next. Strategy has to include a deliberate handoff from "prove it works" to "own it in production," with a named owner and a budget line that exists before the pilot even starts, not one that gets requested afterward.

None of this argues against investing in AI. The direction of the spending is not the problem — the trillions of dollars flowing into this space reflect a real and lasting shift in how work gets done. The problem is that most companies are investing in AI the way they'd invest in software: buy the tool, roll it out, measure usage. The organizations closing the adoption-investment gap are the ones treating AI as an operating-model change that happens to involve software, not a software rollout that happens to involve AI. That distinction is the whole strategy.

AI Strategy Roadmap

Step One: Get comfortable with being curious. 

Secure leaders ask questions and seek to understand. Insecure leaders won't ask questions for fear of being discovered as "lacking" in some way. There are very few true experts in this space right now — make it your goal to build AI fluency and make intentional decisions as you move through your innovation journey.

Step Two: Understand your personal readiness, and your organization's AI readiness. 

To support this step, Servant created a free assessment designed to provide a baseline, inform planning, and mitigate the risk of assuming a readiness level that may not be accurate. 

Step Three: Define AI use guidelines. 

These should reflect your core values, your culture, and your leadership posture on the use of technology — and can be revisited as your organization's AI maturity grows. For example, one key design principle Servant often recommends to clients is this: AI is to be used to manage processes and tasks so human effort can be invested in work that requires the human touch. We aren't replacing people — we are freeing them to do higher-level, meaningful work.

Step Four: Prioritize effort.

 In alignment with your firm's strategic goals and AI use guidelines, decide the problem or problems you want AI to help you solve. This typically falls into one of two buckets: grow revenue (earned or donated), or reduce cost. Once you settle on a specific business objective, define success metrics, assign a leader with the authority to drive change, and report on measurable impact on a regular cadence. The technology leader is a key partner in building AI maturity and executing the plan, but the greatest success comes when business unit leaders choose the problem worth solving and a cross-functional team is assembled to solve it together.

Step Five: Go. Measure. Recalibrate. Go again. 

The greatest risk for enterprises today is doing nothing — falling prey to analysis paralysis. This is a moment for real leadership courage. Don't wait until every condition is perfect. Instead, find a meaningful pilot, scope it, test the hypothesis, learn, adjust, and move forward.

Why This Matters for Faith and Flourishing Leaders

There is one more reason this matters for those of us serving the faith and flourishing ecosystem. The organizations in this space operate under constraints most enterprises do not. When a ministry, a nonprofit, a university, or a mission-driven business spends a year on a pilot that never leaves proof of concept, that is not a line item written off at the end of a quarter. It is capacity that was owed to the people they serve. That reality is not an argument for moving slower, but it is an argument for moving deliberately. Choose the problem worth solving. Name the owner. Fund the integration, not just the tool. Do that, and AI stops being an expense you have to defend and becomes what it should be: time and attention returned to the work that only a human being can do. That is the return worth measuring, and it is the one this ecosystem exists to deliver.

About the Author

Melissa Peak is the Chief Growth Officer at Servant (Servant.io), a strategy consulting firm created to enable Faith and Flourishing leaders to grow, thrive, and flourish in the digital era. Previously she held leadership positions at Amazon Web Services, Pearson Education, and Apollo Education Group, where she helped clients drive transformational change and innovate. She studied at MIT Sloan School of Management, focused on Strategy and Innovation, and serves on the Board of Trustees for Indiana Wesleyan University and Austin Christian University.