Catalyst: Why Legacy Operating Models are Slowing AI-Enabled Transformation 

As part of our 2026 Catalysts research, we’re exploring how organizations are progressing on their AI journeys through Prophet’s Human-Centered Transformation Model. Rather than looking at AI adoption through a technology lens alone, the framework examines how strategy (DNA), capabilities (Mind), operating models (Body), and behavior (Soul) must evolve together to enable lasting transformation. Read the first three articles in our Catalyst series: 

• AI’s Next Challenge Isn’t Adoption, It’s Leadership
• Getting More Value out of AI Means Looking Beyond Tech
• AI Adoption Reshapes Traditional Apprenticeship and Expert Learning Model

Here we focus specifically on our findings under ‘Body’. Organizations are discovering that AI cannot scale effectively on top of fragmented systems, siloed decision-making, and operating models built for a different era. What begins as a technology initiative quickly becomes a broader challenge of governance, prioritization and process design. In other words, AI is exposing the strengths and weaknesses of how organizations operate. The next wave of AI-enabled value moves beyond introducing new tools, to redesigning how work gets done, creating the conditions for experimentation to scale, and building operating models that allow AI-enabled ways of working to take hold. 

What we’re seeing from the new research:

Pilot proliferation is creating noise instead of scale.

Many organizations are running multiple AI pilots simultaneously, often led by different functions, teams, or business units. While experimentation remains essential, leaders are increasingly concerned about duplication, competing priorities, and a growing inability to scale successful initiatives. Without a clear enterprise-wide view, organizations risk creating more noise than value. 

AI individual tactical solutions, one trick ponies, are data environment expensive and not scalable.

Business Leader, Financial Services

AI is exposing organizational debt.  

AI is acting like a spotlight on long-standing operational roadblocks. Fragmented processes, unclear ownership, siloed data, and inconsistent ways of working are becoming more visible as organizations attempt to embed AI into day-to-day operations. 

Scaling AI requires redesign, not redeployment.

Organizations are increasingly recognizing that AI transformation cannot be achieved by layering new technology onto old operating models. The greatest barriers are often structural rather than technical. New technology requires new infrastructure: process, governance models, performance systems, and decision-making frameworks all need to evolve alongside the technology. 

It ends as a productivity tool if that’s how you use it, not a transformative tool.

Business Leader, Financial Services

What’s next for business leaders?

1. Create Enterprise-Level Visibility of AI Initiatives

Our 2025 research emphasized empowering teams to experiment with AI. That remains critical, but experimentation without visibility creates fragmentation. Organizations need a shared view of AI initiatives to reduce duplication, accelerate learning, and scale what works across the enterprise.

2. Align AI Investment Through Prioritization

As the volume of AI initiatives grows, prioritization is becoming more than an operational exercise. Organizations need clear prioritization frameworks to decide what to scale, what to stop, and where AI creates the greatest business value, not just the greatest excitement.

3. Redesign the Workflow, not Just Work

Perhaps the most significant shift from last year’s research is the realization that AI transformation requires operating model transformation. Organizations need to redesign workflows, decision-making, and supporting systems around AI-enabled work, rather than layering AI onto legacy ways of operating.