Everyone Has ChatGPT. Only Some Are Getting Faster.

Usage is standard. ROI is measured. The constraint shifted. Leaders who build internal capability while competitors buy seats will own the next twelve months.

GenAI is now everyday work. The winners will fix the people bottleneck.

I’ve been in enough exec meetings this year to know the feeling. Lots of pilots. Lots of screenshots. Not enough proof that this thing is paying rent. The new 2025 Wharton & GBK AI Adoption Report cuts through the noise. In short: GenAI has moved from “interesting” to standard practice, ROI is now measured, and the next constraint isn’t tools, it’s people.

Why this matters now

Usage isn’t fringe anymore. Nearly half of leaders say they use GenAI every day, and more than four in five use it weekly. IT and Procurement are out front, while Marketing and Operations lag. The sector split is clear: Tech, Finance, and Professional Services lead; Retail and Manufacturing trail, even though the use cases are obvious. That tells you two things. First, your competitors are already turning this into muscle memory. Second, the variance is cultural and process-driven, not just a tooling gap.

The work itself is very practical. The top performers are repeatable knowledge tasks you already fund. Think data analysis, summarizing meetings and documents, writing and editing, presentations, and fast ideation. Adoption and satisfaction line up in these categories, which is rare. This is where the time comes back and where the perception of value hardens into habit.

What’s actually changing inside firms

Accountability arrived. A strong majority of enterprises now track GenAI ROI. About three in four report positive returns, with the fastest wins in digital, process-heavy sectors. Retail is slower, which matches what we see when physical operations create handoffs and data friction. Leaders aren’t waiting for perfection; they’re wiring results to throughput and profit, not just usage charts.

Budgets are moving up, but with discipline. Most leaders expect to increase spend next year, and a meaningful share of technology budgets now goes to internal R&D. Translation: leaders aren’t just buying seats. They’re building custom capability on top of their systems and data. Early agents are showing up too, mostly with a human in the loop for process automation, analytics, internal support, and customer service. Autonomy later. Throughput now.

The people bottleneck

Here’s the uncomfortable part. While most leaders say GenAI enhances skills, many also see a risk of proficiency decline without deliberate practice and coaching, especially for junior roles. It’s the “use it or lose it” effect. Training investment is softening even as hiring pressure rises. CAIO responsibilities exist in most enterprises, often added to existing leaders rather than as stand-alone roles. The tech is ready and governance is maturing, but capability building is uneven and, in some places, sliding backward.

Policies are catching up as access broadens. A large majority of firms now allow org-wide usage and are tightening guardrails on data security, compliance, and oversight. Teams are even using GenAI to manage risk itself, from IT security to financial risk scenarios. This is the right order of operations. Open access with clear rails beats shadow tools and fragmented risk every time.

What this means for you

If you’re a senior leader, stop treating GenAI as an app. Treat it as capacity. Start where the work already proves out: analysis, summarization, writing, presentations, and ideation. Put simple, shared measures on those tasks. Hours returned, cycle time, error rates, and, where it applies, revenue started. Make ROI tracking someone’s job, not a slide. The firms showing gains are the ones reporting against throughput and profit, not just activity.

Then fix the human side. Decide where you will train, where you will hire, and where you will buy help. Don’t underfund practice. A one-time workshop won’t protect skills. Give teams recurring reps on real work and coach to quality. Pair that with open access and clear guardrails so managers say yes with confidence.

Finally, choose where to build. With a growing share of budgets going to internal R&D, pick two or three workflows worth customizing. Aim for high-volume, rules-heavy work that sits near revenue or risk. Build small, ship fast, and put a number on the outcome.

I’ll say it plainly. The gap this year won’t be who has ChatGPT or Copilot. Everyone does. The gap will be who turns everyday usage into measurable outcomes while keeping skills sharp. That’s a leadership problem. And it’s solvable.

---

Business leaders are drowning in AI hype but starving for answers about what actually works for their companies. We translate AI complexity into clear, business-specific strategies with proven ROI, so you know exactly what to implement, how to train your team, and what results to expect.

Contact: steve@intelligencebyintent.com