ZDNET’s key takeaways
- IT modernization has been happening for many years.
- AI brokers might add cohesion to modernization efforts.
- Solely 10% of IT chiefs use agentic AI for modernization.
In some unspecified time in the future in your profession, or at many factors, you may need been requested to assist with an IT modernization program, porting functions from older programs, comparable to mainframes, to newer programs, be they on-premises or in the cloud.
Additionally: 5 ways to use AI to modernize your legacy systems
Modernization has been on CIO agendas for many years. Well-liked efforts have included internet companies, microservices, cloud, virtualization, and x86-based Home windows and Linux servers. These efforts aimed to create extra superior and versatile programs.
When reflecting on these tendencies, it typically looks like modernization is at all times one thing that’s 5 years away. Now, it’s AI’s turn to modernize IT systems.
Enter agentic AI
The outcomes of modernization, as we’ve recognized it, have been lower than stellar. Kyndryl researchers discovered that modernization is commonly simply one other costly, time-consuming venture.
“Solely about half of these modernizing report reaching higher IT operations, and fewer than half are seeing innovation positive factors,” the examine’s authors noticed.
“Practically one in 5, 18%, say their companies noticed restricted or unclear worth from modernization efforts.” As well as, nearly half of respondents stated they’re not on time and struggling value overruns.
Additionally: The best AI leaders are ‘a little bit off the wall,’ says Google Cloud executive
Modernization should turn into greater than swapping out mainframes or other legacy systems for one thing sleeker.
Enter agentic AI, which may form the following wave of modernization. This time, as a substitute of changing mainframes and different legacy programs, the main focus is on increasing tangible enterprise capabilities.
“Price discount and legacy escape are not the main motivations to modernize,” reported Kyndryl. “As AI will increase demand on practically each ingredient of IT, and as sovereignty issues turn into extra acute, leaders need their complicated know-how estates — a mixture of outdated and new selections — to work higher.”
Additionally: 12 rules of agentic AI for successful enterprise transformation
Brokers can function autonomously whereas offering extra visibility on dependencies between programs, each legacy and new. “Fewer than one in 10 organizations are absolutely assured they know these dependencies immediately, a major technical blind spot,” the Kyndryl report acknowledged.
“Agentic AI is displaying early promise to deal with longstanding challenges in modernization and pace progress,” the Kyndryl authors added.
“Agentic AI is getting used to map hidden dependencies, generate code, and create documentation — all historically time-consuming, error-prone processes. Organizations not utilizing agentic AI for modernization are more likely to be not on time than early adopters.”
Tomorrow’s legacy
Nonetheless, modernization challenges stay regardless of AI’s guarantees. Even essentially the most forward-looking organizations can discover modernization to be an limitless quest.
“IT modernization more and more appears to be like like a power situation,” agreed Andy Thurai, founding father of The Discipline CTO and former chief strategist with IBM, commenting on the examine.
“In keeping with this report, 4 in ten enterprises are nonetheless working mainframes. We maintain listening to modernization is the number-one mandate, but the end line is at all times 5 to 10 years out.”
Additionally: AI agent kill switch urged by Okta-led alliance — how businesses could make it work
One other situation is that “organizations are addicted to purchasing software program,” stated Thurai. “The legacy programs beneath it nearly by no means get decommissioned. AI is now the most recent shiny distraction from the grind of untangling that technical debt.”
CIOs should take care of a state of affairs through which “vendors are pitching AI to IT leaders as the cure-all,” Thurai continued. “The economics don’t maintain up. When AI drives infrastructure sprawl, the month-to-month invoice turns into a black field. The report discovered that three in 10 leaders already fear AI-driven modernization will finish in wildly unpredictable compute prices.”
Additionally: What workers are really using AI for — and what they aren’t
AI-based modernization might have its second within the solar, particularly for organizations “that went previous the bolt-on chatbot,” stated Thurai. “They’re utilizing agentic frameworks as an orchestration layer to map out hidden technical dependencies, after which producing the lacking documentation as they go. Agentic AI is rushing up initiatives that used to eat months of error-prone handbook labor.”
In brief, brokers might be simply the most recent chapter of a nonstop journey. “In the end, immediately’s AI-driven architectures are going to be the legacy environments someone else has to modernize,” stated Thurai.
