Reframing Software Testing in the light of AI

W14     Start Time : 14:30     End Time : 16:45

Talk Abstract:

This workshop gives a firm grounding in what AI is, and what AI is not. We explore the deceptive nature of some AI, and the social and personal consequences of poorly designed and implemented AI. AI means that software is moving from deterministic outcomes, to a variable range of outcomes, which adds additional complexities, and responsibilities, on testing in the new future of AI. This tutorial gives you the knowledge and techniques to evolve your practice and update your skill set.

We cover two main impacts of AI on testing: testing AI-based applications, and new AI tools and techniques for testing. We show you how to address reliability versus validity issues in software testing of AI systems, and give you a deep and detailed understanding of how algorithmic and data biases affect systems and their testing. Will AI-based automation tools be a software developers best friend? We review the different approaches taken to using AI in testing, and how they vary in enabling speedy, exhaustive, and realistic tests. We then give you the skills to assess what you need to consider before selecting or using AI automation tools, and what you need to know to evaluate and effectively integrate these tools in your practice.

In summary, AI is changing the nature of software, as well as the human processes involved in developing it, as immense, rich, and complex data sets become an intrinsic part of the systems we develop. This tutorial will give you the tools you need to improve the efficiency and precision of your testing practice.

Key Takeaways:

  1. An understanding of how AI is transforming software development, and how software testing is changing in the light of AI
  2. How to use AI-based testing tools to improve the effectiveness of your practice
  3. How to address reliability and validity, and algorithmic and data biases, when testing AI applications

Back to Programme

    • Speaker


      Chris McKillop - CEO/Co-Founder, Turalt, Canada

      Dr. Chris McKillop is a thought leader on AI, ethics, and empathy. She is CEO/Co-Founder of Turalt. Dr. McKillop has a degree in Artificial Intelligence, an MSc in Human-Centred Computer Systems, and a PhD in Educational Technology which focussed on the emotional dimensions of learning. Her research led to over 25 publications and numerous international conference presentations. She has a long record pioneering technological innovations to improve learning in both academic and workplace settings. Her personal mission is to show how AI can help us understand our humanity, and to help make the world a more empathetic place.