Code Intelligence, a leader in AI-driven software testing, has announced the launch of Spark, the first AI-powered test agent capable of autonomously identifying vulnerabilities in unknown code without requiring human input. This groundbreaking tool has already proven its capabilities by discovering a real-world vulnerability in widely used open-source software through automated test generation and execution.
Revolutionizing Software Testing with AI
Spark is designed to fully automate the software testing process, from early bug detection during development to remediation. By lowering the barrier to adopting advanced security testing methods like white-box fuzz testing, Spark enables teams to enhance code quality and security while significantly reducing manual effort.
For a codebase of 100,000 lines of code, Spark can save up to 1,000 hours of manual testing, making it an invaluable asset for development teams striving for efficiency and accuracy.
Real-World Impact: Spark Uncovers Vulnerability in WolfSSL
During its beta testing phase, Spark uncovered a heap-based use-after-free vulnerability in WolfSSL, a popular open-source cryptography library commonly used in embedded devices and IoT systems. The AI agent autonomously analyzed the code, generated a test case, and identified the vulnerability without human intervention, requiring only a single command to activate the process.
This vulnerability, if exploited, could lead to unexpected behavior, crashes, or security breaches. The WolfSSL team promptly addressed the issue, releasing a patched version in December 2024.
Also Read: DeepEdge.ai Partners with Ambarella for Next-Gen Edge AI at CES 2025
“The uncovered real-world vulnerability proves that AI can effectively take over manual tasks in software testing, such as analyzing code, identifying the most likely attack vectors, generating and running tests, and can thereby yield great results,” said Eric Brueggemann, CEO of Code Intelligence. “Next, we will focus on going even further by automatically fixing any uncovered bugs. This means the entire software testing process—from creating tests to bug remediation—will be completed in minutes without human interaction. However, humans will continue to make the final decisions. We will provide automatically generated pull requests with a proven fix.”
Empowering Developers with Advanced Testing Capabilities
Spark seamlessly integrates with development workflows, simplifying the traditionally complex and resource-intensive process of fuzz testing. Its capabilities have already impressed industry professionals.
“We were truly impressed by the abilities of Spark to enhance our fuzz testing workflows,” said Andreas Lackner, Senior Software Development Engineer at Vector Informatik. “By reducing the manual effort for creating and integrating fuzz tests, we are able to bring our cycle time down and further improve the quality of our embedded software.”
The Future of AI-Driven Software Testing
With Spark, Code Intelligence is setting a new standard in automated software testing. By combining cutting-edge AI with proven testing methodologies, Spark not only streamlines bug detection but also promises to revolutionize remediation in future iterations, further empowering developers and security teams to ensure robust, secure software.
