Rewiring for AI: From ambition to advantage

In this episode of the McKinsey Podcast, host Lucia Raheli speaks with McKinsey partners Rob Levin and Kate Smage about their updated book, "Rewired: Protecting Value, Generating Value, and Leading Through the AI Revolution." The conversation centers on how companies can achieve success in the age of AI, emphasizing the shift from isolated pilot projects to company-wide, end-to-end workflow transformations. Rob Levin highlights that the second edition of "Rewired" was prompted by the rapid evolution of AI, particularly generative AI and agentic AI, which have fundamentally changed capabilities like software development. The core thesis of the book remains relevant: companies that have built foundational AI capabilities are better positioned to succeed in this new AI landscape. He uses Freeport-McMoRan as an example, a company that achieved success with its initial AI investments in digital twins and then leveraged generative AI for further improvements in its leaching process. This illustrates the need for continuous adaptation and developing a "second muscle" for flexibility in the face of technological change. Kate Smage elaborates on the tangible value capture from AI transformation, citing research on a cohort of 20 companies that have successfully implemented the "Rewired" framework. These companies have seen an average EBITDA uplift of 20%, become cash-accretive within one to two years, and achieve a return of three dollars for every dollar invested. She emphasizes that these successes are often achieved by focusing on a limited number of high-impact domains (three or fewer) rather than spreading resources thinly across the organization. Rob outlines the framework for successful AI transformation, which involves a set of enabling capabilities: aligning on a business-led roadmap, focusing on key domains, developing talent, adapting the operating model, ensuring high-quality data, and prioritizing adoption and scaling. He notes that with agentic AI, the focus has shifted to reimagining end-to-end workflows rather than just individual use cases. Furthermore, the need to consider the entire workforce, not just scarce tech talent, is crucial, as agentic AI is poised to impact all white-collar workflows. The disruption in software development, with tools like Cloud Code, offers significant productivity gains but also introduces complexity in solutioning technology. Kate adds that successful companies operate at a different "metabolic rate," enabling them to move faster from insight to action. DBS is presented as a case study, where foundational AI investments allowed them to quickly capitalize on generative AI, resulting in significant verifiable benefits. She likens this capability to a muscle that strengthens over time, allowing for continuous acceleration. When asked about getting started, Kate suggests focusing on "domain change" – targeting areas of greatest economic leverage within an industry – and prioritizing talent density. She stresses that AI transformation is fundamentally a people transformation, requiring leaders to consider how human and AI employees will collaborate. Rob highlights speed as a defining organizational advantage, referencing DBS's ability to deploy models much faster than before. He also points to "agentic engineering" as a key area, citing LATAM Airlines' advanced adoption for the entire software development lifecycle. Regarding accountability, both Rob and Kate agree that AI transformation must be a top priority for the CEO and a distributed effort across the entire leadership team. It's a "corporate team sport," not something to be delegated solely to the IT department. CHROs, CFOs, and business owners must all be actively involved. Rob identifies common gaps in AI maturity, including the misconception that AI is purely an IT initiative rather than a business-led one, and a lack of preparation for adoption and scaling beyond the MVP stage. He stresses the importance of resourcing for full-scale deployment, efficient technology "kitting" for reuse, and addressing the non-technical aspects of adoption, such as engaging with external partners. Finally, Kate advises building AI conviction by focusing on value and solving real business problems. She also encourages leaders to be forgiving of the process, acknowledging that the cost of iteration is decreasing, making it easier to pivot from wrong turns. The emphasis is shifting from finding the perfect answer to stress-testing, owning, and building conviction around solutions.