GFT Analysis Says AI Documentation Can Cut Maintenance Work 30%
A GFT Technologies analysis says AI-linked software documentation can cut maintenance effort and speed developer onboarding when knowledge assets stay synchronized with code changes.
The Economic Times carried an ANI story on a GFT Technologies analysis that moves software documentation into the AI development workflow, after enterprises measured maintenance-work reductions as high as 30% and developer onboarding gains as high as 40% when system knowledge stayed current.
The finding moves documentation away from its usual place at the end of a coding cycle.
When notes, runbooks and diagrams are written after the software is finished, they can fall behind as code changes, leaving new developers to decode existing systems and maintenance teams to trace dependencies without a current map.
GFT's analysis treats that lag as an operating problem rather than a housekeeping task.
AI tools can read code structures, dependencies and logic, then turn those relationships into explanations of what a component does and how it connects to the rest of an application.
A payment processing change shows the mechanism.
If a developer updates that module, AI can refresh the related API documentation, sequence diagrams and runbooks alongside the code change.
The documentation then follows the latest version of the software instead of waiting for a separate manual update.
That shift matters most in large or older systems, where developers can spend significant time searching for how existing code and processes work.
GFT's own analysis put current enterprise adoption above 65% for either documentation tasks or code analysis; its claimed efficiency case depends on knowledge assets tracking system changes before onboarding and maintenance improve.
The same evidence points to a governance use case.
If the generated material shows dependencies, code logic and component relationships, maintenance teams get a clearer path through older applications and regulated businesses get a more current explanation of how important systems operate.
Andre Gagne, CEO of GFT Technologies Canada, framed the change as a broader efficiency layer across the software lifecycle.
He said a 30% maintenance reduction is only part of the value because automatically updated documentation can help financial institutions show regulators how critical applications work and can support modernization projects that depend on understanding legacy systems.
The analysis also sets limits around automated documentation.
AI-generated material still needs validation and monitoring, including controlled version histories, auditable change records, authenticated access and clear disclosure of how the content was produced.
The practical result is a workflow change: documentation becomes a live part of software delivery, rather than a separate record created after development work is done, and the claimed payoff remains anchored in the same 40% onboarding and 30% maintenance-effort measures.
