Rewired to win: Reimagining the enterprise with tech and AI

The discussion centers on the challenges and strategies for companies to successfully leverage Artificial Intelligence (AI), particularly generative AI, for business value, as detailed in the revised edition of the book "Rewired." The core thesis is that while identifying AI's potential value (the "what") is becoming more standardized across industries, the true differentiation lies in a company's capability to execute and scale AI initiatives (the "how"). Rob Levin and Kate Smage, co-authors of "Rewired," emphasize that the rapid evolution of AI, from machine learning to generative and agentic AI, necessitated an updated framework. The second edition addresses the disruption of software development by AI and the imperative for established companies to adapt their operating models, leadership, and capabilities to not just compete but to "out-innovate, out-run, and outperform." A key takeaway is that companies that built foundational AI capabilities in the "AI 1.0" era are better positioned for the "AI 2.0" landscape. Freeport-McMoRan is cited as an example, having successfully applied its existing capabilities to new areas like leaching after initial success with a digital twin of its copper concentrator. This demonstrates the compounding value of established AI competencies. The scale of AI-enabled transformation is significant, but the research presented indicates it is worthwhile. A cohort of 20 companies rigorously applying the "Rewired" framework has achieved an average EBITDA uplift of 20%. Importantly, this value is not a distant promise; these companies become cash accretive within one to two years, with two-thirds achieving this with three or fewer targeted domains. The return on investment is also compelling, with an average of $3 generated for every $1 invested. This contrasts with companies that attempt to implement AI broadly without the necessary capabilities, often falling short of these impressive results. The "Rewired" framework is built on six core capabilities: aligning on a business-led roadmap, focusing on key domains, developing enabling capabilities (talent, operating model, control functions), optimizing the tech stack and data, and prioritizing adoption and scaling. While these foundational elements remain crucial, the "timely" aspects have evolved significantly with generative AI. In strategy, the emphasis has shifted from individual use cases to reimagining entire end-to-end workflows. Agentic AI's ability to automate large portions of workflows makes this holistic approach essential for success. McKinsey's research shows that companies thinking end-to-end, rather than simply integrating AI tools into existing processes, are far more likely to realize value. On talent, the focus has broadened from scarce technical talent to encompass the entire workforce. Agentic AI is poised to automate many white-collar workflows, necessitating a strategic approach to workforce transition, skill development, and redefining roles in collaboration with AI. Technologically, the 20x productivity gains in software development, driven by generative AI coding assistants, are collapsing traditional development team structures. However, this era also presents increased complexity in vendor selection and potential for high ongoing operational expenses. A thoughtful approach to technology architecture, avoiding brittle or inefficient point solutions, is critical. Finally, adoption remains a significant challenge. Reinventing end-to-end workflows requires a clean-sheet approach, involving task automation, role reconstruction, and extensive training. This magnitude of change management, often neglected, is crucial for successful scaling. "Rewired" companies, according to Kate Smage, operate at a different "metabolic rate," characterized by reduced latency between insight, decision, and action. This speed allows them to out-compete. DBS Bank is highlighted again as an example, moving from 18 months to production for its first AI model to deploying one every two months. This agility, built on foundational capabilities, enables compounding value and continuous acceleration, akin to developing a muscle rather than completing a one-time transformation. For leaders seeking to initiate or accelerate their AI journey, practical next steps are distilled into signature moves. Key among these is focusing on "domains" – areas of greatest economic leverage – rather than scattering resources across numerous individual use cases. Targeting three or fewer impactful domains, such as forecasting in retail or claims processing in insurance, is recommended. Another critical element is "talent density," recognizing that AI transformation is fundamentally a people transformation. Ensuring a strong talent proposition and preparing for the integration of human and "silicon" employees is paramount. Rob Levin emphasizes the importance of speed as a defining organizational advantage, citing DBS's dramatic reduction in model deployment time. He also highlights "agentic engineering" as a fundamental disruption, enabling rapid development across the entire software lifecycle, as seen with LATAM Airlines. This shift also underscores the importance of domain expertise complementing technical talent, as precise definitions of "done" can be translated into code by AI. Accountability for AI transformation is framed as a "corporate team sport." While CEO sponsorship is non-negotiable, ownership must be distributed across the entire leadership team. The Chief Human Resources Officer, Chief Financial Officer, and business domain owners all have critical roles to play, moving beyond the traditional delegation of AI to the IT department. Common missteps include the failure to lead transformations with business objectives rather than solely relying on IT to execute. Companies often fall back into old paradigms of working with IT, which are insufficient for the current pace and scope of AI change. Another significant gap is the lack of preparation for adoption. This includes under-resourcing the scaling of successful MVPs, neglecting efficient technology "kitting" for broader deployment, and underestimating the non-technical aspects of adoption, such as engaging external partners like suppliers. Building AI conviction, both individually and organizationally, is crucial before the window to lead closes. The best way to foster conviction is by focusing on tangible value and solving real business problems. Leaders are also encouraged to acknowledge the inherent difficulty of this work and to leverage the reduced cost of iteration offered by agentic AI to test, pivot, and build confidence. The process is often messy, and valuable insights can emerge from what doesn't make it to production. The discussion concludes by urging leaders to assess their own rewiring efforts using a provided assessment tool and to consult further resources on McKinsey's website. The importance of continuous adaptation in the age of constant change is underscored.