Lambdatest Inc.a software testing platform powered by generative artificial intelligence, today announced the release of its private beta model. Testing from AI Agent to Agent A service that allows developers to validate and evaluate AI agents.
With the short time that AI agents have become a trend, they have become a major aspect of digital transformation. Unlike traditional AI chatbots, AI agents can take autonomous actions by providing goals by breaking down human surveillance into step-by-step tasks that require little or no human surveillance.
As a result, their behavior is adaptive and unpredictable. That is, there is no standard way to test them to ensure reliability.
“Every AI agent you deploy is unique and it's both its greatest strength and its greatest risk,” says Asad Khan, co-founder and CEO of Lambdatest. “As AI applications become more complex, traditional testing approaches cannot keep up with the dynamic nature of AI agents.”
The rise of Agent AI It also means that more enterprise developers are working themselves with the complex network of interacting agents more The system and Each other. Agents can interact with people more than just texts. They can make voice calls and read PDFs, Please take a look Watch images, videos and interact with your computer screen.
The lambda test said so that By allowing the team to upload requirements documents in various formats, such as text, images, and videos, we have developed a proofing platform for AI agents to handle these scenarios. The system generates test scenarios by processing this analysis and simulating real-world challenges break AI Agent Under stress.
The platform highlights important metrics such as bias, integrity, and hallucination. Using these criteria allows teams to better analyze agents' quality and meet specific needs. company I said system Included 15 dedicated test agents, ranging from security researchers to compliance variators.
For example, Agent AI could be deployed in customer service roles that require you to maintain a casual tone with your customers. It also needs to interoperate with other agents who get information about stocked items and provide quick responses to order updates. Tests may include how well it is included Provided Brand service, how quickly it responds, its reliability and security.
Reliability and security are a major concern for businesses. a 2025 Ernst & Young Global Ltd.'s investigation revealed 73% of senior leaders believe that one day the entire business unit could be managed by Agent AI. Still, this future is hampered by barriers like cybersecurity, data privacy and corporate policy management.
Unlike a single-factor test system, agent-to-agent agents act as the “brain” that AI agents use for inference and test generation. This multi-agent approach creates a more comprehensive and detailed test suite, providing teams with a wide variety of alternatives and potential edge cases for AI applications.
“Our platform thinks of us as real users who generate smart, contextual test scenarios that mimic real-world situations where AI can struggle,” says Khan.
Image: Siliconangle/Microsoft Designer
By engaging with the TheCube community, we support our mission to keep content open and free. Join TheCube's Alumni Trust Networktechnology leaders connect, share intelligence and create opportunities.
- Over 15m viewers of TheCube Videothe power of conversations such as AI, cloud, cybersecurity and more.
- 11.4k+ TheCube Alumni – Connect with over 11,400 technology and business leaders who shape the future through a unique, trustworthy network.
About Silicon Angle Media
Tech Visionarys Founded by John Furrier and Dave Velante, Siliconangle Media has created a dynamic ecosystem of industry-leading digital media brands reaching 15 million elite technical experts. Our new, unique TheCube AI Video Cloud is based on audience interactions. Leveraging TheCubeai.com neural networks, we help technology companies make data-driven decisions and stay at the forefront of industry conversations.
