Organizations in the Asia-Pacific region are increasingly sacrificing software quality for speed, a trend Damien Wong, regional senior vice president at Tricentis, described as “scary.”
In an interview with Singapore's Computer Weekly, Wong cited the company's recent research on software quality transformation, which found that almost half of the organizations surveyed admitted to releasing code without testing it.
“The pressure to deliver quickly sometimes leads to cutting corners and releasing software before it has been thoroughly tested,” Wong said. “Companies have a huge need for speed.”
Untested code can have significant financial consequences. According to Wong, software defects and outages can cost organizations an average of $500,000 to $5 million.
To address this, Tricentis has invested heavily in agent-driven test automation. Unlike previous artificial intelligence (AI)-powered software test assistants that function as co-pilots, AI agents have a degree of autonomy, allowing them to generate test assets, run them, and analyze results with minimal human intervention.
“Imagine having a virtual performance engineer and being able to instruct them using natural language,” Wong says. “We can say, 'Test your system with up to 1,000 concurrent users, return the results, and make some recommendations to improve performance.'”
Wong pointed out that using automated software testing tools will not put human testers out of work. In fact, you'll be more productive and valued much more than you were before. “Imagine having one highly talented AI-enabled test engineer who can do the work of 10 or 20 engineers to handle the through-the-ceiling speed of code development.”
Demand for automated software testing tools is reflected in the company's regional results. Wong revealed that Tricentis' Asia-Pacific business grew 42% year-on-year in fiscal 2024, outpacing the company's global growth rate of 28%.
The company's growth has been driven by two key trends in the region: application modernization and generative AI-driven digital innovation. As traditional “monolithic and brittle” applications are updated to cloud-native stacks, testing complexity will increase exponentially, Wong said.
“There are hundreds, thousands, even tens of thousands of applications running across the organization,” he said. “One change to one system or application can have potential impacts on other systems or applications upstream or downstream.”
Wong noted that visibility is also important in agent testing, and in the broader software quality assurance realm. He pointed to the company's acquisition of SeaLights, an AI-powered software quality intelligence platform, and explained how AI can prevent disasters similar to the 2024 CrowdStrike failure caused by content configuration updates that were not thoroughly tested.
Wong said the SeaLights platform can constantly monitor code changes in application metadata and correlate them with testing activities. “For example, if you change the API [application programming interface] Have you checked in your code and run what kinds of tests? If no tests have been done, it will intelligently tell you that this is a problem. You won't be caught off guard if something goes wrong. ”
As organizations rush to integrate generative AI into their products, they are also grappling with how to validate non-deterministic large-scale language models that produce different outputs from the same input, making traditional validation methods inadequate.
“There is no right or wrong outcome, but something can happen that is completely unacceptable. So we need to have clear guardrails and make sure we stay within them,” Wong said.
Although the market is still maturing in this area, Tricentis' recent masterclass on testing intelligent systems in Singapore was oversubscribed, indicating that organizations are concerned about AI risk and compliance, Wong said. But for now, Wong said, most customers are using AI to test traditional deterministic systems.
Tricentis tools are available through the public cloud and on-premises. The company has established cloud regions in Australia and Japan, and plans to do so in Singapore, but Wong noted that many regulated industries still prefer on-premises or hybrid solutions.
“Trends are mixed in the Asia-Pacific region,” he said. “There has been a big trend towards fully public cloud migration, but there has been a bit of a setback.Many of the highly regulated customers we work with are choosing to deploy our software on-premises due to security concerns,” he added.
That said, Wong assured customers who use Tricentis' cloud platform that the company does not have access to the source code. “We can look at the metadata associated with a build of software and determine which modules have changed, but we can't know exactly what code is inside. So we can protect people from the fear that their code will be stolen or stolen,” he said.
