AI in Software Test Automation

Stanislav Dinev, Software Test Architect at Musala Soft
Nowadays we all observe the vast speed of tremendously evolving technologies, more and more of which are facilitated by the Artificial Intelligence at its finest.
AI consist of two inevitably connected major parts:
Machine learning
and
Taking decisions (proposing an educated guessed opinion in different spheres of life) based on using really optimized algorithms which work over the big collected and classified statistical (and best practices) data sets.
Stepping even further we can observe that we already have an AI creating basic applications by facilitating the developers automate the routine code creation (code skeletons), as well as proposing better code design patters, refactoring and optimizations over an existing application code.
The main lecture topic here concerns one layer above - the software automation testing and how AI would facilitate it by helping creating test data sets, propose decisions and test strategies, help make the test results reporting better and even creating test automation code skeletons such as Page Object Models based on the DOM of a certain page.
Comparison of the main automation tools using AI, as well as how well they self educate and become better and better, (expectations versus reality) is also covered.