I am new to AI Driven Testing (AI-DT). I've explored few tools like testim.io, endtest.io. Is any other open source tools are there to achieve this?
First of all, I doubt that there is currently something like AI-driven testing.
*-driven means that your entire development/testing process strongly relies on
*. Working for retest myself, I'm convinced—due to my experience there—that AI can be a great help to complement traditional testing approaches. But: Today's AI is inferior to hand-crafted test scripts when it comes to test generation. (At least for now.)
Testim, for instance, incorporates historical data to rank the locators for each element individually to stabilize the tests over time. This is awesome and may lead to a comeback of capture & replay. However, you still create tests manually since a human decides what to test and how. Therefore, I wouldn't speak of AI-driven testing.
To get back to your actual question: When you say you tried out Testim and Endtest, I assume you want to test web apps? Unfortunately, I'm not aware of similar open-source solutions, but I think it's just a matter of time until it's developed by the community on top of tools such as Selenium. Existing open source testing tools that use AI are e.g. EvoSuite (unit test generation) and Infer (static code analysis).
Do you use Selenium? Klaus (ex-colleague) currently works on a side project in his spare time to extend selenium tests with AI. He is looking for contributors and testers. Maybe you are interested to join him?
Information about AI-driven Testing Tools is consolidated in the AIFORSE Landscape, but there're no known Solutions which would be open source. It's a bit early for this.
P.S.: I'm a part of the AIFORSE Community, but it is not affiliated with any of the Tools listed in the Landscape.
I am not aware of anything open-source at this time or even an open source project. Given how many years it took us at Appvance to achieve machine generated tests (6 years and a team of 30 people) using 19 different AI algorithms, it would seem a daunting task for an open source project. We are at about 4 million lines of code and growing.
That said, AI generated tests can reduce the need to script by 90% or more, and provide close to 100% coverage of many, but not all, applications.
While I know everyone wants free tools, there is real ROI and savings moving to AI generated tests that far exceeds the cost of the technology. People costs in QA outweigh technology costs. It's possible that some software, with excellent support, is worth paying for.
Disclaimer - I work for Appvance.ai