AI's role in modernizing aging IT systems faces economic doubts
AI is being promoted as a solution for aging IT systems, but a former IBM strategist questions whether the financial case works. A Kyndryl survey of 2,000 senior IT decision-makers found that only 10% use agentic AI for modernization. Many modernization efforts still suffer delays, cost overruns, and limited returns, according to the article.
A former IBM strategist argues the financial rationale for using AI to overhaul legacy IT is questionable. Vendors market AI as a broad remedy, yet a Kyndryl survey of 2,000 senior IT decision-makers found just 10% apply agentic AI to modernization. Earlier waves—web services, microservices, cloud, virtualization, and x86 servers—aimed at more flexible systems, but results have often disappointed.
Kyndryl’s findings show about half of modernizers report improved IT operations, fewer than half see innovation gains, and 18% find limited or unclear value. Nearly half are late or over budget. Agentic AI may help by mapping hidden dependencies, generating code, and producing documentation, though fewer than one in ten organizations feel fully confident about those dependencies.
If AI modernization remains limited, organizations and public-facing services relying on older systems could face continued delays, higher costs, and uneven service quality. IT workers may see changing roles, with dependency mapping and documentation tasks increasingly automated. Taxpayers, customers, and employees might bear consequences of costly upgrades with unclear returns. Conversely, successful agentic AI adoption could improve reliability and innovation, though benefits may arrive slowly and unevenly.