Gen AI as a software quality tool? The skepticism is fading as more organizations implement it

Is artificial intelligence (AI) the answer – or at least a partial answer – to nagging software quality problems? Software quality has been a challenge since the first computers were built eighty years ago, and in a world overrun by technological networks and solutions, the problem has only become more acute. A new study suggests that generative AI (gen AI) is emerging as an important step in quality management.

According to a study by Capgemini and Sogeti (part of the Capgemini Group), which surveyed 1,755 tech executives, there is increasing emphasis on incorporating gen AI within quality engineering. 68% of organizations use gen AI to support their quality efforts. At least 29% of organizations have fully integrated gen-AI into their test automation processes, while 42% are actively exploring its capabilities.

Also: Google’s new AI course teaches you to write more effective prompts – in 5 steps

“The evolution of large language models and AI tools, especially Copilot, has enabled their seamless integration into existing software development lifecycles, ushering in a new wave of efficiency and innovation in quality automation,” says the team of authors of the investigation, led by Jeff Spevacek. from OpenText, mentioned.

In last year’s Software Quality Survey, “we saw an increase in organizations’ investments in AI solutions to drive the quality transformation agenda,” Spevacek and his co-authors wrote. “However, a significant number were skeptical about the value of AI in quality engineering.”

Also: Do ​​you think AI can solve all your business problems? Apple’s new research shows otherwise

Furthermore, attitudes towards AI have changed significantly over the past 12 months, she added. “A large number of organizations are now moving from experimentation to large-scale deployment of gen AI to support high-value engineering activities. We truly believe we will see further progress in this area.”

However, deploying AI as a software quality assurance tool is not without its challenges. At least 61% of respondents are concerned about data breaches associated with the use of generative AI solutions. A lack of comprehensive test automation strategies and reliance on legacy systems were identified as major barriers to advancing automation efforts by 57% and 64% of respondents, respectively.

Also: Can AI make data science obsolete?

Some of the OpenText/Sogeti team’s recommendations for advancing automation and AI in software quality efforts include:

• View company-wide: Clearly define “the objectives and desired outcomes of high-performance engineering automation and pre-select the areas where test automation should be applied, increased, or improved.”
• Start now and keep experimenting: “If you are not yet exploring or actively using gen-AI solutions, it is critical to start now to stay competitive. Don’t rush to commit to a single platform or use case. Instead, experiment with multiple approaches to identify those that provide the best solutions. main benefits.”
• Leverage the full range of Generation AI capabilities: “Gen AI goes far beyond the generation of automated test scripts and helps realize self-adaptive test automation systems.”
• Connect the company’s key performance indicators: “Identify and leverage the key business performance indicators impacted by high-quality engineering automation, with a clear focus on business outcomes such as increased customer satisfaction, reduced operating costs and others relevant to the business.”
• Rationalize high-performance technical automation tools: “Ensure your high-performance engineering automation tools are streamlined and integrate with emerging technologies, such as generation AI, to maintain compatibility and future readiness.”
• Enhance high-quality technical talent and roles: “Include more full-stack quality and software development engineers in the tests to strengthen your team’s capabilities.”
• Improve, not replace: “Understand that Gen AI will not replace your quality engineers, but will significantly increase their productivity. However, these improvements will not be immediate; allow sufficient time for the benefits to become apparent.”

Software quality engineering is evolving rapidly, the authors pointed out. “Once defined as testing human-written software, it has now evolved to include AI-generated code.” Quality engineering increases the number of code and test scripts to be generated, as well as requirements for end-to-end testing of software chains.