DevOps platform company Perforce Software has laid down a new AI validation tool. The software service features a new capability within its Perfecto continuous testing platform for web and mobile applications. Aiming to create a new way organisations conduct software testing, especially within the realm of enterprise applications that make use of AI models and/or Large Language Model (LLM) technologies, Perforce says its approach avoids the need to create multiple scripts and frameworks, which are often cumbersome and typically tough to scale on across multiple platforms.
AI fire to fight fire
Aiming to provide an AI validation service that moves autonomous testing closer to reality through context-aware testing, Perforce says its service uses advanced artificial intelligence to validate applications visually and contextually and dynamically adapts to application changes without human intervention.
According to Stephen Feloney, vice president of product management at Perforce, many existing testing solutions rely on AI co-pilots to simply create more scripts, but Perforce’s AI validation tool uses natural language prompts and does not rely on objects or code, instead creating durable tests that work across platforms.
He calls it a “user-centric approach” that eliminates the need for specialised scripting knowledge so that more users can adopt and scale test automation quickly. By removing co-pilot complexity, Perforce moves towards autonomous testing, an AI-driven approach to testing that eliminates the need for human intervention.
“The success that our early adopters have already experienced with AI Validation is a huge validator to our approach with testing,” said Feloney. “Creating more frameworks and more code in co-pilot does not help testers do what they have always wanted: validate exactly what appears on the screen. This is what AI Validation provides them. Perfecto’s AI validation service is aligned for enterprise teams navigating multi-platform testing with integration into CI/CD workflows to enable continuous and scalable testing.”
Object locator flakiness
By simplifying processes and enhancing adaptability, Perforce says it enables teams to maintain quality and speed across the development life cycles. Perforce Perfecto client, Midwest Tape, a physical and digital media distributor, has incorporated the technology into its testing strategy, reducing their overall testing time by 20%. The company has said that because its application base relies heavily on [object] locators, which can often be unreliable and prone to flakiness, the use of AI-driven validation significantly enhances stability and efficiency.
“We are excited about where this can go,” said Byron Chan, digital delivery quality assurance Lead at Servus Credit Union, a US-based financial institution. “I see tremendous potential because eventually, you could come up with test cases in this prompt format before development even starts and then once developed/deployed, you could potentially avoid manual testing and automation test development because it’s already done.”
Whether validating a complex trending graph, bar chart, or a dynamic calendar view – across Android, iOS and varied screen resolutions – Perforce says its AI validation improves quality, lowers maintenance and reduces costs while changing how testing is done.
Unlike basic OCR-based solutions, Perfecto’s AI-driven approach verifies the meaning behind dynamic elements – charts, dashboards, or graphics – to ensure user experiences reflect the intended content. This deep level of coverage ensures thorough visual validation across all application layers.
Move over manual muscle
The company reminds us that manually updating scripts or object locators whenever an application changes can lead to frequent test failures and more maintenance. Perfecto’s AI Validation inherently avoids this issue by eliminating reliance on fragile locators and script updates – so there is no need for continuous adjustments when the UI evolves.
Stopping short of talking about the “democratisation of AI testing” and the potential for “citizen AI testers” to now evolve, Perforce has at least said openly that “anyone can test” with these new services; the company explains that while test creation and maintenance demands specialised scripting skills, limiting participation to a few technical experts… AI validation’s natural language prompts open testing to everyone on the team, expanding coverage while freeing specialists to tackle more complex challenges.
