AI dramatically reshapes enterprise software engineering teams

Forty percent of organizations report that artificial intelligence generates the majority of the code merged into production within the last 90 days, according to The Futurum Group. This widespread adoption means AI is increasingly responsible for core enterprise systems, fundamentally altering development workflows. More than half, 54%, of organizations currently use AI across more than half of their software development lifecycle, indicating AI's deep integration into product pipelines.

However, this rapid output is leading to a significant increase in production incidents and codebases that human engineers cannot fully understand or maintain. The speed of AI-driven development is creating a disconnect between code generation and human oversight, complicating error resolution and future enhancements.

Companies are currently trading immediate development velocity for long-term operational stability and maintainability, a trade-off that will likely result in escalating technical debt and unforeseen system failures.

Who is AI Impacting in Enterprise Software Engineering?

• Seventy-five percent of organizations have experienced a production incident where AI-generated code, AI agents, or AI tooling was a contributing factor, according to The Futurum Group.
• Enterprise software engineering teams are increasingly relying on AI, as 40% of organizations report AI generates the majority of their production code, according to The Futurum Group.
• Companies integrating AI deeply into their Software Development Lifecycle (SDLC) face new challenges, with 54% of organizations using AI across more than half of their SDLC, according to The Futurum Group.
• The Software Lifecycle Engineering (SLE) market is projected to grow from $111.1 billion in 2025 to $226.0 billion by 2030, according to The Futurum Group, indicating new investment needs.
• Engineers maintaining AI-generated code face difficulties because AI ships code at machine speed, but the understanding of the code does not ship with it, making AI-assisted codebases harder to maintain and evolve, according to Infoq.
• Traditional SaaS providers and investors are affected by changing market perceptions, as Wall Street investors believed CIOs would use AI to replace SaaS solutions, driving the "SaaSpocalypse" prediction, according to CIO.