The state of the art in automated software testing is far from being a replacement for human-guided testing. There is more to testing than setting up preconditions, applying inputs, verifying outputs, and logging the results.
Testing requires significant planning, exploring, learning, modeling, inferencing, experimenting, and more. Therefore, before we can truly automate testing, we must bridge the gap between the testing capabilities of humans and machines. Tariq King says that breakthroughs in artificial intelligence (AI) and machine learning (ML) are challenging our thinking about the types of problems that machines can tackle. Can AI discoveries—a machine that masters a game like Go or autonomously drives an unmanned vehicle—help us find better solutions for automated oracles, test generation, system modeling, and defect discovery? Tariq believes they can and will share his vision of how. Drawing on his experiences working on, leading, and advising teams in the development of software that automatically tests software, Tariq walks us through recent advances in AI and ML. Join Tariq as he maps these advances to potential solutions for important software testing research problems.