AI can cut software maintenance effort by 30%, speed developer onboarding: Report

Artificial intelligence (AI) is changing how companies manage software documentation, with teams reporting up to 30% lower maintenance effort and up to 40% faster onboarding when technical information is kept updated with changing systems, according to an analysis report by GFT Technologies.

It said software documentation, which has traditionally been treated as an end-stage task, is increasingly becoming a continuous part of the development process.

In traditional software development, documentation is often prepared after coding is completed. This can leave information incomplete or outdated as the software changes, making it harder for new developers to understand systems and for maintenance teams to identify how different parts of the software work.


It said, "For years, documentation has been the weakest link in the software development lifecycle. ...Today, that is changing fast. AI-driven documentation is turning what used to be a burden into a continuous, intelligent process that supports development speed, quality and transparency".

According to the company, AI can understand code structures, dependencies and logic and use that information to explain not only what a piece of code does but also how different components are connected.

For example, when a developer changes a payment processing module, AI can update related API documentation, sequence diagrams and runbooks along with the code change. This allows documentation to remain aligned with the latest version of the software.

The shift could be particularly relevant for companies managing large or older software systems, where developers can spend significant time looking for information about existing code and system processes.

The company said "over 65% of enterprises already use AI for documentation or code analysis. Teams report up to 40% faster onboarding and 30% lower maintenance effort when knowledge assets stay in sync with evolving systems".

Andre Gagne, CEO of GFT Technologies Canada, said the data reflects a shift in how companies are using AI in software development.

"AI stops being a productivity experiment and becomes a foundation of efficiency across the software lifecycle. A 30% reduction in maintenance effort is only part of the story. When documentation stays current automatically, financial institutions can demonstrate with confidence to regulators how their critical applications work. It is also a key success factor for any institution starting a modernisation journey; you can't modernise what you don't understand," Gagne said.

The company said AI-generated documentation still needs to be validated and monitored. It highlighted the importance of version control, audit trails, secure authentication and transparency around how documentation is generated.

As AI becomes more widely used across software development, documentation is increasingly being linked directly to the development workflow rather than being treated as a separate task.