Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
AI may speed up parts of legacy IT modernization—especially mapping dependencies, generating code and producing documentation—but the available evidence does not show that it can fix legacy systems on its own or that AI-led modernization reliably pays for itself. The central question is not only whether AI can help do the work; it is whether the resulting systems, operating burden and recurring infrastructure costs make economic sense.
What AI is being used for in legacy modernization
Early reported uses of agentic AI in modernization include mapping hidden technical dependencies, generating code and creating documentation. These tasks can be labor-intensive, and better visibility into how applications and systems connect could help teams plan changes. But these are reported use cases, not evidence that AI can safely replace experienced engineers, validate every dependency or deliver a successful migration without review.
That distinction matters because legacy environments often contain undocumented relationships and business rules. AI-generated maps, code and documentation should be treated as inputs for technical validation, not as proof that a system is ready to change.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
How widely is agentic AI being applied?
ZDNET reported that 10% of technology chiefs in a Kyndryl survey of 2,000 senior IT decision-makers said they were applying agentic AI as a modernization tool. The article also reported that almost half of respondents said modernization efforts were behind schedule and experiencing cost overruns; 18% saw limited or unclear value from modernization, and fewer than one in ten organizations were fully confident they knew their system dependencies. These are figures attributed to Kyndryl material in the article, not independently validated benchmarks. The article does not provide the survey field dates, detailed sample composition or exact question wording, so the results should not be generalized beyond that account. ZDNET’s article
#1 Best Overall
Why the economics may not work
Andy Thurai, founder of The Field CTO and a former IBM chief strategist, argues that AI can make infrastructure sprawl and its recurring costs harder to understand. “The economics don’t hold up. When AI drives infrastructure sprawl, the monthly bill becomes a black box.” That is an attributed concern, not a quantified cost study, but it points to a practical risk: a lower-effort coding or discovery task does not automatically make the full modernization program cheaper.
Teams need to consider transition costs alongside ongoing compute, software and operating expenses. If AI adds services or infrastructure without simplifying the environment or retiring older systems, the organization may pay to run both the new and legacy estate. Thurai’s broader criticism is that modernization can perpetuate complexity when organizations keep adding technology without removing what came before. He described the pattern this way: “IT modernization increasingly looks like a chronic condition.”
What the reported findings do—and do not—say
The article also quotes Kyndryl’s report: “Cost reduction and legacy escape are no longer the leading motivations to modernize.” That statement describes reported motivations; it does not establish whether modernization succeeds, whether AI improves results, or what return organizations achieve. Likewise, the survey figures do not compare AI-assisted projects with conventional ones or provide a controlled measure of savings.
So, will AI make a difference where earlier modernization efforts have fallen short? It may help with discrete discovery and documentation work, but the evidence presented does not establish that AI resolves the organizational, technical or financial causes of stalled programs. Nor does it show that an AI tool can make the hard decisions about which systems to change, how to manage risk or what to retire.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How to judge an AI modernization proposal
Before committing to an AI-led approach, compare proposals against the whole lifecycle rather than a narrow demonstration of code generation or dependency mapping. The following are evaluation criteria, not outcomes established by the reported survey:
- Total cost: Include transition work and recurring infrastructure, software and operating expenses.
- Schedule and overrun risk: Ask what assumptions drive the timeline and how changes in scope or hidden dependencies affect it.
- Dependency visibility: Require teams to validate discovered relationships against the systems and business processes they support.
- Code and documentation quality: Define how outputs will be reviewed, tested, maintained and kept current.
- Security and governance: Establish who can approve generated changes and how access, data handling and auditability will be controlled.
- Retirement plan: Identify which older applications or infrastructure will actually be decommissioned, and when.
A proposal that demonstrates faster output on one task but leaves these questions unanswered has not yet shown that it makes modernization economical.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
