Companies Must Fix Old Tech Before AI Can Really Work, Expert Says

The leader of a large technology company called Capgemini says that businesses around the world want to use artificial intelligence, but most of them are not ready yet. On Thursday, July 30, Capgemini's chief executive Aiman Ezzat explained that companies will need to spend years fixing and upgrading their old computer systems before AI can truly work for them. He said this need for upgrades will create a long stretch of investment and growth for the technology industry.

Ezzat said the biggest problem holding back AI is not a shortage of AI tools or programs. Instead, the real obstacle is old technology that companies have been using for decades without replacing. These aging systems often keep data stored in messy, disconnected places, which makes it very hard for AI to find the information it needs to do its job.

He used the word 'agentic' to describe AI that can handle many steps of a task all on its own. Ezzat said every company wants that kind of powerful AI, but they first have to make their systems 'AI-ready.' Right now, he said, most companies are simply not there yet, even if they have been trying to use AI tools.

Capgemini calls this period of fixing and upgrading a 'modernization supercycle.' That means the company expects businesses to spend heavily on improving their data systems, software, and main technology infrastructure for many years to come. Capgemini itself is doing well because of this trend, and it recently raised its goal for how much money it expects to earn in 2026.

One big reason companies are behind is something called 'technical debt.' This happens when businesses take shortcuts with their technology over many years, leaving systems that do not work well together. As a result, data gets scattered across systems that cannot share information, and AI tools struggle to get what they need.

Even newer AI tools that can answer questions still have trouble carrying out full business tasks when the systems underneath them are broken or disconnected. Ezzat said that companies are now spending their money more carefully on AI. Instead of running small experiments, they are choosing bigger, more complete technology projects that can change how their entire business works.