The Truth About AI and Testing: There Is No Debate
At a recent testing conference there was the familiar session, or rather debate, on when AI generated tests would arrive and their impact. Since I was not invited to the debate stage (it was instead vendors with only legacy scripting offerings), the debate raged on with little regard to the facts. That it would be many years or never, and how AI might be used or not.
Since we invented and launched AI driven testing in 2017 (see patent), Appvance IQ has been used to generate millions of regression tests autonomously, and blueprint hundreds of applications, many every day.
Is AI coming for your job?
Will it find bugs we otherwise would not find?
Does AI help us meet DevOps compressed time schedules?
Is now the time to embrace AI?
Will it work with my website or application?
Here is what we have learned so far:
No, AI is not coming for your job. Real AI (at Appvance) is being used to dramatically increase code and test coverage. We see clients at 8% today achieve over 90% coverage in a matter of weeks. There will still be scripts for specific use cases (and you can create them 20x faster in Appvance Test Designer than Selenium). So…scripting doesn’t go away. But the reliance on scripts for coverage does fade away. Instead scripts are used to prove specific user flow business logic.
Because the coverage is so high, AI finds bugs at UX and API levels you would not otherwise find easily or at all. And that is the primary goal of QA to begin with. Find more bugs without spending more time or money. AI fulfills this goal nicely and makes management very happy!
With AI generated tests, and highly parallel testing, we see QA times fall from weeks or days to hours. The tests created are data-driven and mimic what your users actually do. You can finally meet the goal of DevOps with rapid test generation and high coverage. And with AIQ, no dangerous keystroke analyzers or trackers need to be added to your code (security won’t let you do that anyway). Nothing is added at all.
Is now the time to embrace AI in your organization? The simple answer is yes, if you have budget and need. The need to move fast (how about creating 5000 tests in 10 minutes?), and the need to find more bugs with less effort. AIQ starts at $100K/yr depending on number of applications, usage and such. It is available on-prem or in the cloud. In comparison to any other method of achieving 20X productivity increase and identifying more bugs earlier in the cycle, the ROI is often 10X or more.
Will it work with my website or application? If it is HTML5 including REACT, Angular etc, yes! AIQ is designed with pure JS to interact perfectly even with modern web apps. And we offer a free evaluation against your application to demonstrate full compatibility before you buy.
Are you ready for your free evaluation? Request time with a technical consultant below and we will get right back to you!