Despite heavy investment and hype, most companies are not seeing real returns from generative AI. New cross-industry research points to a gap between ambition and execution. The findings suggest that business results arrive when leaders change how they guide teams, not just what tools they buy.
The study highlights a small set of firms that are pulling ahead. Their leaders act as “shapers” who connect technical work to business goals, build trust, and apply AI responsibly. The timing matters. As budgets tighten and scrutiny grows, companies want proof of value, not proofs of concept.
Why Many AI Efforts Stall
Executives are under pressure to show outcomes. Yet many projects begin with a model, rather than a problem worth solving. Teams then struggle to scale pilots, train staff, and manage risk. Without clear leadership, these hurdles turn promising demos into sunk costs.
Researchers report that the “overwhelming majority” of generative AI programs have not delivered “significant business returns.” That blunt assessment reflects a pattern. Strategy lags behind experimentation. Governance trails deployment. Adoption stalls because people do not trust the tools or do not see how they help their work.
The Shaper Approach
“The small percentage of companies who actually are unlocking AI’s potential are distinguished by leaders who act as ‘shapers,’ aligning technical innovation with business strategy, fostering trust, driving adoption, and embedding AI responsibly into operations.”
Shapers do not run AI as a side project. They treat it as a business change that touches process, talent, and culture. They frame clear use cases, tie them to measurable goals, and set expectations for risk and performance. They also talk about limits, which builds credibility.
The research identifies five behaviors that appear again and again among these leaders:
- Strategic agility: Move fast, but keep a steady link to business needs.
- Human centricity: Design with workers and customers in mind, not only the model.
- Applied curiosity: Test, learn, and share lessons across teams.
- Performance drive: Set targets, track outcomes, and course-correct quickly.
- Ethical stewardship: Set rules for safety, fairness, and accountability.
From Champion to Capability
“Rather than concentrating authority in a single AI champion, successful firms need to intentionally cultivate these skills in leaders across teams and levels.”
Many organizations name a single AI lead and expect results. The research warns against this structure. It can slow decisions and create bottlenecks. It also isolates expertise.
Shaper firms spread leadership skills across product, engineering, risk, HR, and operations. This shared approach shortens feedback loops. It also raises the odds that adoption sticks because managers closest to the work can guide change.
Trust, Adoption, and Responsible Use
Employee trust is the make-or-break factor. People adopt tools that help them deliver better work with less friction. They resist tools that add steps or feel risky. Leaders who invite input, explain decisions, and measure impact see faster uptake.
“Leadership will become an accelerant, not a barrier, to AI transformation.”
Responsible use is part of that trust. Ethical stewardship means clear guardrails. It includes controls for accuracy, privacy, and bias. It also clarifies who is accountable when things go wrong. Teams are more likely to try new tools when they know the rules.
What To Watch Next
Results will follow where these leadership behaviors take root. Expect companies to shift spending from pilots to scaled solutions with hard metrics. Look for more training for managers, not only data teams. Also watch for tighter alignment between AI goals and business results, such as cycle time, revenue, or customer satisfaction.
The main takeaway is simple. Technology alone does not deliver returns. Leaders do. Firms that build shaper skills across levels will cut waste, raise trust, and scale what works. Others will keep chasing demos while value stays out of reach.
