Semafor World Economy

AI is poised to become the most transformative general-purpose technology since electrification, reshaping operations, decision-making, and value creation across industries.

Companies are deploying AI throughout their supply chains, customer interfaces, and internal workflows, seeking both efficiency gains and entirely new capabilities. The gap between early adopters and slower movers is widening, driven by differences in data readiness, talent, governance, and executive alignment.

Scaling AI requires rethinking organizational design, risk management, and the relationships between human and machine decision-making. Firms must determine who is accountable for outcomes as AI systems take on more complex roles. The challenge lies not only in adopting tools, but in integrating them into the core fabric of the enterprise.

Incumbent businesses are in a heated race against small, AI-native companies, with the legacy companies falling behind.

Large, established companies are at an inherent disadvantage in the era of AI. Many are saddled with technical debt, have outdated Enterprise Resource Planning systems, and rely on code bases that humans wrote by hand.

For them, taking no risks is not an option.

That's because these scrappier AI-native companies are being built differently. Workflows are created around the use of AI agents. Humans don’t write any code, and the code that they approve is built so that it can be updated and changed with AI.

To compete, legacy companies should look at where they have advantages. For instance, years of data can be leveraged with AI, but it takes work and investment. Companies that think of AI as a cost-saving measure, finding opportunities to cut jobs and or slow hiring, are thinking of the AI moment incorrectly. The AI transition, ironically, requires a lot of human labor. The moment requires investment, rather than cost savings.

For years, AI was treated as an emerging capability. Important, but deferrable. That assumption no longer holds. From what I see across enterprises today, AI is now a determinant of competitiveness and long‑term value creation. The question is no longer if AI will matter, but whether organizations are prepared to make it work at scale.

Our research and direct client experience show a widening divide. A small group of enterprises has moved beyond experimentation and is delivering measurable value. Many others remain stalled. The difference is not technology, it is readiness. Where AI has failed to scale, it has exposed familiar weaknesses: fragmented data, unclear governance, outdated operating models, and workforces unprepared for human‑machine collaboration. Treating AI as an add‑on rather than embedding it into how the enterprise runs has proven to be a costly mistake.

Those seeing sustained returns made harder choices early. They invested in data discipline, governance, infrastructure, and workforce readiness, often before outcomes were fully visible. In some cases, they changed how decisions are made and how accountability works. That shift was uncomfortable, but it is what separates experimentation from advantage.

At Infosys, we partner with enterprises to operationalize AI responsibly and at scale, shifting the focus from deploying tools to embedding AI into the fabric of the business. This is now a leadership responsibility and a prerequisite for sustained shareholder value.