A practical guide to testing AI-Based Systems
The stories we are fed about AI seem to fall into two categories:
1)It’s our saviour! It’s going to be wonderful!
2)It’s going to take all our jobs and destroy humanity!
The media feed us so many simplified stories designed to stimulate our imaginations with science fiction fantasies, both utopian and dystopian. The law and governance around AI is also so far behind the curve that it can confuse the issues, often propagating the misconceptions and making life harder for those wanting to use AI sensibly. The truth is far more nuanced and requires us to understand much more than just the hype. Whether we like it or not, AI is here to stay and as Testers it is our job to understand those nuances, to understand what the actual risks are and how to mitigate them.
So ...How do you test an AI system? This talk addresses some of the risks, opportunities and ethics involved. It provides a framework for testing & assuring AI-based systems. It also explores the technology itself, along with the techniques, skills sets and thinking patterns that we as Testers need to adopt to remain relevant, as the fourth Industrial Revolution kicks in for real.
AI is evolving. We need to as well!
