Software program testing firm Code Intelligence in the present day introduced Spark, an AI check agent that may establish bugs and vulnerabilities with out human interplay.
In its beta interval, as an illustration, Spark discovered a vulnerability in WolfSSL the place the one human involvement was working a single command to launch the agent. The agent dealt with analyzing the code, producing a check case, and working the check case all by itself.
“The uncovered real-world vulnerability proves that AI can successfully take over guide duties in software program testing, corresponding to analyzing code, figuring out the almost certainly assault vectors, producing and working exams, and might thereby yield nice outcomes,” mentioned Eric Brueggemann, CEO of Code Intelligence.”
With Spark, Code Intelligence hopes to totally automate software program testing, from figuring out bugs to remediating them. It believes its answer will decrease the barrier to entry for superior testing methods.
In line with Code Intelligence, Spark can save as much as 1,000 hours of guide effort for codebases with 100,000 strains of code.
The corporate might be holding an official launch occasion for Spark on January 28 at 9 AM EST, the place it’ll go over extra particulars concerning the new answer.