AI Transformation Lessons

Eighty-eight percent of organizations now use AI in at least one business function. As of Q2 2026, only 1% describe their rollout as mature. That gap — between broad experimentation and real enterprise value — is where most transformations stall.

If your pilots are generating headlines but not P&L impact, this is written for you.

The window is open, but it will not stay open indefinitely. Companies that lead with AI are growing revenue up to 3.7 times faster than laggards, and 73% of CEOs plan to increase AI investment over the next twelve months. The advantage is real, and it is compounding. The question is no longer whether to move. It is whether you will build the discipline to convert motion into value before the window narrows.

Having spent the last several years inside one of the industry's largest AI transformations, and having led four prior technology paradigm shifts, I have come to a simple conclusion. AI transformation fails or succeeds on the same fundamentals that have always separated real transformation from expensive pilots: strategy, leadership, product operations, and high-performing teams. AI simply raises the stakes and compresses the timeline. The AI paradigm shift is a business and workforce transformation.

Why Most AI Transformations Stall

Gartner's AI Hype Cycle names the pattern well: a Peak of Inflated Expectations followed by a Trough of Disillusionment stalls progress. Organizations rush into proofs of concept, generate early buzz, and then confront a hard truth.

A working model is not the same thing as a working business process.

The research converges on why. The World Economic Forum, MIT, Accenture, PwC, and McKinsey all trace the stalling and lack of value realization of GenAI and Agentic AI to the same source: bolting AI onto existing workflows, instead of redesigning how work actually gets done. Many companies are implementing AI ad hoc, resulting in fragmented delivery and suboptimal results.

In our experience, redesigning workflows and business operations are critical, but how to identify and execute is where most AI transformation initiatives stall, missing one or more of these three capabilities:

The Lessons That Matter Most

I have read hundreds of research papers, playbooks, books, and frameworks, and was a leader for one of the largest AI transformations while learning and adapting along the way. Much of the published guidance now converges on the same ideas. I'm honored to have been one of the first to build and execute this approach in practice, and it's validating to see the research land in the same place. Based on my experience, here are the lessons that matter most.

Strategy and Economics

Create and execute a comprehensive AI enterprise strategy and organizational change management plan, with executive leadership named, outcomes defined, and instrumentation in place. The experimentation phase is ending. Be intentional about the journey.

Stay alert to AI economics. They are shifting rapidly and will impact your ROI and EBITDA.

Redesign or augment, and automate your value streams, workflows, and processes where it is financially rewarding and technologically feasible. Be prepared to stop, pivot, or expand.

Operating Model and AI Fluency

Organize teams around products aligned to customers, and their supporting platforms and services. Embrace the product operating model to accelerate integration and benefit realization across an interconnected enterprise.

Evolve your IT and engineering approach for the AI era: incremental discovery and delivery, service management integrated for production-ready AI at scale. Agile is not dead, it is evolving.

Prepare for the human-machine workforce. By 2030, agentic AI is widely expected to work alongside people as digital employees. Grow your human talent and retain institutional knowledge.

Data, Governance, and Advantage

Treat data as a product and your IP moat. Consider building your own models; model sovereignty will become a competitive advantage.

Get AI governance and cybersecurity right early, and be ready to adapt as the landscape shifts.

Avoid long-term vendor lock-in. The pace of technology change is unpredictable.

The Project Brilliant Path: Diagnose, Strategize, Scale

The organizations pulling ahead follow a disciplined, iterative path rather than a single big-bang rollout. The same discipline shows up across McKinsey's Steer-Scale-Institutionalize model, Gartner's crawl-walk-run guidance, and BCG and MIT Sloan's research on AI maturity. At Project Brilliant, we quickly apply that discipline across three activities.

The 1% Solution

AI is not another technology transformation or adoption program. It is a once-in-a-generation opportunity to redefine how your enterprise competes, operates, innovates, and grows.

The organizations that lead this shift will not simply deploy new technology. They will reinvent how decisions are made, how work gets done, how talent creates value, and how markets are shaped. The future will belong to leaders who move with clarity, conviction, and speed.

Start with the 1%

Transformation does not begin everywhere at once. It begins with the critical 1%: the right leaders aligned around the right priorities, making the first bold decisions that unlock the other 99%. Project Brilliant helps leaders find that 1%, act on it, and turn it into enterprise-wide transformation.

Project Brilliant Recognition: I thank Grace Sondermann, Andy Lien, Diana Williams, and Aaron Kopel for their ideas, editing, and collaboration.

References