The art of pivoting

I attended the Responsible AI summit (day 1) today. It has been very interesting about the governance aspects. As I reflect on the day I do wonder if there needs to be more bridges between governance and technical — in particular testing efforts. There was a huge emphasis on risk and also lots of talk of testing, however, how are the two linked? As a testing professional my core focus is on risk and value. We have a technique called risk based testing where we focus our testing efforts based on our evaluation of risk. Or if we’re talking about use cases we could look at persona or role based testing. Overall, I thought that testing leaders need to be in the room and bridge gaps between the business and technical side of things. Testing is going to be a core part of evaluating AI products and testing expertise in this area will be vital in leading testing efforts of AI products.

I also thought in some cases testing may have been misinterpreted. There was a slide that attempted to explain AI Assurance and it described testing as a check execution of a system under controlled conditions. Testing for me isn’t about that — the word that they also used was evaluation, which seemed more appropriate. Making a judgment based on the information you have in front of you with normally doing some type of comparison with the product under test against standards, risks or whatever the mental comparison model may be. It’s also about continuous testing — testing throughout from inception to monitoring in production.

I also reflect on the AI lifecycle and its importance in guiding the overall process of building AI products. If you’d like to learn more on that then it’s worth looking into the ISO/IEC 5338:2023. It’s not only what we do but it’s going to be how we go about it. What is the process of getting from A to B? If team members are moving in different directions then it will become problematic.

My overall reflection though is the use of the word pivot. To use what we already know and adapt the strategy, approach, technique or principle towards AI. Continuous testing pre and post deployment. Asking very good questions like “does it make sense?” and so on.

After today I for one am excited about my career in the software testing arena. It is a very exciting field and I hope one day to lead an organisation on AI Testing Strategy by using all my transferable skills and knowledge (along with new knowledge gained from attending conferences like today.)

If there is ever a time to pivot. The time is now.