Follow ZDNET: Add us as a preferred source<!–> on Google.
ZDNET’s key takeaways
- Many IT execs are concerned AI is creating more tech debt, not less.
- AI may be running on debt-laden platforms.
- Longer-term thinking is needed for software.
Tech debt has long been an ever-expanding tax on technology implementations, and it appears that the problem is unlikely to go away anytime soon – even if AI takes on some of the heavy lifting in application development and deployment. In fact, AI may make things worse.
A new study, published jointly by HFS Research and Unqork, finds that 43% of IT managers fear AI will create new technical debt – even as 84% expect cost reductions from AI adoption.
Also: AI may take jobs, but it makes starting a business easier than ever – here’s how
The survey, conducted in September, involved 123 executives and managers from large companies. There are high hopes that AI will help cut into and clear up issues, along with cost reduction. At least 80% expect productivity gains, and 55% anticipate AI will help reduce technical debt.
However, the large segment expecting AI to increase technical debt reflects “real anxiety about security, legacy integration, and black-box behavior as AI scales across the stack,” the researchers indicated. Top concerns include security vulnerabilities (59%), legacy integration complexity (50%), and loss of visibility (42%).
What is technical debt? This occurs when technology teams, usually under pressure, apply quick fixes or shortcuts to software issues, versus more deliberate projects. The result is a need for more costly rework and maintenance in the future to eventually unravel the initial shortcuts.
“Technical debt exists at many different levels of the technology stack,” Gary Hoberman, CEO of Unqork, told ZDNET. “You can have the best 10X engineer or the best AI model writing the most beautiful, efficient code ever seen, but that code could still be running on runtimes that are themselves filled with technical debt and security issues. Or they may also be relying on open-source libraries that are no longer supported.”
Also: The key to AI implementation might just be a healthy skepticism – here’s why
For example, one Unqork client is now unraveling 25 years of tech debt incurred through their Java applications, Hoberman related. “Just updating the underlying Java Virtual Machine was so costly and slow – it takes months – with so little benefit. By the time they finished updating one version, a new one would be released and need to be upgraded. That’s the definition of inherited tech debt.”
AI presents a new raft of problems to the tech debt challenge. The rising use of AI-assisted code risks “unintended consequences, such as runaway maintenance costs and increasing tech debt,” Hoberman continued. IT is already overwhelmed with current system maintenance.
4 ways to avoid creating more technical debt
How to guard against creating new forms of tech debt via AI? Here are four recommendations:
1. Tamp down AI projects that rapidly generate code without traceability, rollback, or integration guardrails
The implementation needs to show who did what, when, and why. Without this, the AI-based system “isn’t just fragile, it’s a future liability,” said Hansa Iyengar, practice leader at HFS and co-author of the survey report. “The most debt-prone AI efforts look fast at first, but quietly entrench complexity and make every change harder down the line.”
Also: 8 ways to make responsible AI part of your company’s DNA
2. Oversee a shift in models and architectures
Move to productized outcomes in architectures “that minimize customer code creation, maximize reuse, and embed governance so that AI reduces, rather than creates more tech debt,” said Hoberman.
3. Press upper management on the urgency of longer-term thinking in software development
Emphasize how software spending will deliver business outcomes such as revenue growth. “When you quantify what’s going to run the business versus what’s driving change, it becomes much easier to engage the board in a fact-based conversation about reallocating spend,” said Iyengar.
Also: Forensic vibers wanted – and 10 other new job roles AI could create
4. Modernize legacy systems
Without such modernization, and without changing the underlying architecture, the benefits of AI will be buried. “In that case, AI will generate more tech debt, not less,” said Hoberman.
