Software testing teams have been led to believe that AI will increase development speed, reduce effort, and support better decision-making at scale. However, despite the benefits that AI can bring, implementation will only be successful if AI tests meet quality standards set by humans.
Reliability and dependability are just as important as speed.
a report Title from Leapwork The gap between AI hype and test automation reality We looked at how companies view AI in their testing applications and whether the technology is meeting expectations.
To better understand how AI will impact the testing role, Leapwork conducted a survey of over 300 QA professionals.
The study found that although there was a strong intention to use AI across test teams, adoption of the technology was uneven and had mixed success, impacting testers’ trust in AI.
AI adoption rate in testing
whole, 65% said they are currently using or considering AI in one or more testing activities, but adoption levels also vary by peer group.
According to the survey, 88% of respondents indicated that AI is a priority in their upcoming testing strategy, and 46% indicated that AI is an important or high priority in their organization.
There was debate as to where AI could add the most value, with most (66%) valuing faster and easier test creation, followed by broader coverage across critical systems (38%) and early defect detection (36%).
Currently, respondents estimate that an average of 41% of tests are automated and 59% remain manual.
What are the barriers to adoption?
The majority of concerns about large-scale adoption of AI were related to AI accuracy and quality (54%).
This has created an environment where testers know AI is available but are hesitant to apply it. Now, when we ask teams why they can’t automate more tests, the top answer is “frequent test interruptions,” followed by “difficulty automating the flow of the entire system,” and “time required to update tests.” These combined factors challenge both tester workload capacity and AI reliability.
Interest in AI testing capabilities is high, but confidence in deploying them at scale is low.
Time constraints were a key issue. Testers were often interested in applying AI to their role, but 54% didn’t have the time to experiment and build the confidence to confidently change their current practices.
Other reasons such as system complexity (45%), budget constraints (44%), and lack of skills or expertise (40%) all contributed to challenges.
While only 12.6% of those surveyed said they currently use AI in their core testing activities, 80% said they expected AI to have a positive impact on testing over the next two years. This suggests that testers expect short-term improvements in technology accuracy and quality, driving adoption.
Recommended reading
When participants were asked whether their efforts to maintain application quality had become more difficult in the past 12 months, only 11% felt that the task had become easier, 43% said the difficulty remained the same, and 46% said that the task had become somewhat more difficult.
Today, as the complexity and volume of systems increases, testers have the opportunity to apply AI to reduce their workload.
Currently, we are facing a gap between the hype of AI and the reality of test automation. To bridge the gap between AI hype and testing adoption, you need to improve your skills, allocate time for experimentation, and conduct robust evaluations to gradually introduce AI.
BBuilding trust in AI output requires robust validation that builds lasting trust in testers. This won’t happen overnight.
