The companies that have truly captured economic value from AI do so not because they alone possess particular tools—those tools are widely available—but because of how and how quickly they apply technology to solve large-scale business problems. Building that ability requires new organizational capabilities, which take time to develop and ultimately form a sustainable competitive advantage.
The authors studied 20 companies that consistently generated significant economic value from AI-enabled business transformation. Many appear in the second edition of Rewired: How Leading Companies Win with Technology and AI. These are not headline-grabbing wunderkind tech firms but large organizations that invested years to build technology and AI capabilities to convert tech into business value.
According to the most recent State of AI report, 94 percent of businesses have not yet created meaningful value from AI. The math is unforgiving and contributes to boards’ and top teams’ skepticism about making the level of investment required for successful AI-driven transformation.
A small set of companies, however, achieved radically different outcomes. For each of the companies examined, researchers reviewed transformation road maps and financial outcomes and interviewed select executives to gather more detail. The analysis surfaced six core capabilities that successful organizations build; these capabilities reinforce one another and require time to mature.
Six capabilities and continuous improvement
Common features across the leading firms include: C-suite commitment to AI; platform and data architecture treated strategically; product-oriented, cross-functional teams organized around customer journeys; reusable platform components and APIs; unified data and AI platforms; and repeatable practices for adoption and scaling.
Leaders do not treat early AI wins as end points. Transformation typically unfolds in waves of two to four years. Freeport, for example, achieved its first large-scale AI breakthrough in 2018 with step-change productivity in its copper concentrators; three years later, in 2021, it applied the same capability to achieve another breakthrough in leaching.
These capabilities are not built overnight. The highlighted exemplar companies have been developing them for more than five years and continue to invest.
The C-suite’s role and tech-capable business leaders
A C-suite that understands AI is the most significant driver of success. Top teams must focus AI efforts where they matter and use AI in ways that are competitively differentiated. That requires strategic creativity and the ability to see opportunities others miss.
DBS Bank’s top management spent substantial time in Silicon Valley and with digital natives to learn that high-performing tech organizations are distinguished by capabilities—modern engineering practices, platform-based architectures, data at scale, agile ways of working, and a culture of continuous experimentation—rather than by specific tools. This insight led DBS to reframe its technology operating model and commit multi-year resources to build those capabilities so technology could reliably accelerate speed, innovation, and customer value.
Because AI is embedded in operations and workflows, harvesting value requires leaders who combine deep domain expertise, technology and data understanding, and the ability to orchestrate end-to-end change. When Freeport went all in on its leaching optimization, it appointed a leader with operational AI experience and credibility in processing operations as general manager of the leaching operation; he acted as the domain owner, leading development, integration, and accountability for outcomes.
Across success stories there is always a senior business leader (sometimes a two-in-a-box model with business and technology co-leads) who integrates business, technology, and change management. Of all upskilling programs, developing tech-capable business leaders is among the most strategic and impactful.
Platforms, product teams and data as enterprise assets
Rewired organizations embed tech delivery capabilities in the business to enable faster, more effective innovation cycles and build platform capabilities to maximize reuse.
DBS organizes work around end-to-end customer journeys—such as opening an account, buying a home, or securing small-business financing—aligning cross-functional teams accountable for seamless customer outcomes. The bank also maintains enterprise platforms (payments, customer data, onboarding, credit) that bring together business, technology, and operations and are funded as long-term assets rather than discrete projects. These platforms produce reusable APIs, data assets, and services that scale across the bank.
This distributed operating model lets teams close to the business build, test, and improve iteratively. Although every success story studied had a version of this model, only about 10 percent of companies have adopted it; rearchitecting around a distributed model requires vision and resolve.
LATAM Airlines intentionally built customer-facing capabilities separately from legacy core systems in the cloud with an API-first, modular architecture. The result: components could be reused across channels and geographies and integrated with local payment systems and regulations via abstraction layers without fragmenting the core platform.
Well-architected platforms accelerate product launches, automation, data reuse, AI deployment, resilience, and more; weak platforms slow everything down and add hidden costs through complexity, fragility, long cycle times, and dependence on a few heroic individuals.
Unified data platforms and measurable impact
In 2018 DBS took 15–18 months to develop and deploy AI models because data was siloed and obtaining access and understanding data quality were lengthy tasks. Teams often built one-off pipelines that were not reusable.
DBS therefore invested in a unified data platform and an AI platform, automated data access controls, and made data discoverable through metadata and consumable via data products. It aligned data governance around business-domain leadership to treat data as an enterprise asset.
By 2023 model deployment time fell to two to three months, and the ease of data consumption was central to unlocking an estimated more than 1 billion Singapore dollars (nearly US $772 million) in value generated from AI.
Adoption and scaling are distinct, hard problems
AI creates value only when systems are adopted and scaled. Adoption often fails because adjacent upstream or downstream processes remain unchanged. Freeport, when scaling a pilot that improved copper recovery via leaching, created data and sensor standards so data could be ingested into an existing cloud data platform and modeled consistently across assets.
Scaling requires modular solution architectures and a well-choreographed interplay between central teams and receiving units. Freeport found that roughly 60 percent of an AI system could be reused across plants while 40 percent needed local adaptation; this guided how the central team maintained shared assets and how field teams handled localization.
These adoption and scaling capabilities are learned over time; a company develops an internal success playbook that evolves with new technologies such as agentic AI.
Stages, agentic AI and the path ahead
Leading companies continually improve AI systems—better data, more sophisticated models, real-time responsiveness, deeper orchestration, and richer customer experiences. Capabilities at each stage build on the previous stage; skipping stages is not realistic.
Most profiled companies have largely mastered stage 2 capabilities and are actively building stage 3 capabilities. LATAM, for example, is advanced in agentic software development adoption: it increased effective development capacity by 50 percent while reducing team size by 30–50 percent.
We are in the early innings of stage 3. Agentic capabilities will mature quickly in the coming years, and companies that master them will compound further advantage over peers that do not.
Build the muscles or react
AI is already reshaping industries. The pressing question is whether organizations are deliberately building the capabilities to shape that future or merely reacting to it. The firms pulling ahead are investing now to develop the organizational muscles that allow them to repeatedly convert technology into durable business value.
(Edited by Barr Seitz, editorial director, New York office.)
