Everything starts with a question

3 minute read
Artificial intelligence

With almost three weeks and 4,000 miles between me and module one of the 2022 Oxford Executive Diploma in Artificial Intelligence for Business experience, I have been thinking a lot about our learnings and interactions.

I was impressed with the international diversity of the cohort, coming from 16 different countries and representing both the private and public sectors. I was enthralled by our deep discussions about ethics in artificial intelligence (AI) and amazed by the intriguing use cases our guest speakers brought us. 

Professor Stephen told us we’d come to appreciate the ‘kinds of challenges leaders must face under a state of technological disruption’. After all this, there was one thing I kept coming back to: the beginning. The beginning of thinking about how to use AI to make a giant leap forward. The beginning before money is invested and directions are given. The beginning before hands hit the keyboard.

I now think AI is too often initiated by technologists pushing new technology onto a problem set. ‘We’ve got this new tech; is there a place we can apply it?’ Or poor leadership throwing technology around seeking a solution for an unknown problem. ‘We are sitting on mounds of data; there’s gotta be something valuable in there.’ But one realisation I’ve landed on is that everything starts with a question, not the technology or data in hand. So, to truly unleash AI’s full potential on a critical issue, a leader must begin with an important question.

The question must be at the root of an important issue for you and your organization. It makes little sense to solve something that is of minor importance. Let’s find some examples in real businesses. How about: ‘What products will I need to meet customer demand next month?’ ‘Can we reduce poaching activity?’ or ‘how do I deliver the needed support for refugees at the right time and place?’

These are big, important questions for an organisation. Solving these will positively impact us and our constituents. But for AI to help us here, we need to reframe these questions into prediction questions. For example, ‘what is the forecasted demand for each stock keeping unit one month out?’, ‘where will poachers likely encounter animals today?’ or ‘what kind of support (medical, shelter, food, communications) will the refugees need in this location?’ Next, we’ll seek the data required to make these predictions, augmenting them if needed.

Here’s where the cohort leaned in hard, pausing to reflect on the question and the implications of solving the vital challenge. Is this being done ethically? Are we harming others in the process of helping some? Are we taking a human-centric approach that will instill trust and compassion? Converting organisational challenges into AI-powered prediction questions isn’t necessarily easy or as obvious as the examples provided here. It will take some effort to build up our own AI intuition and reframing techniques. Taking those questions forward with the widest view and ethical considerations requires authentic leadership, which brings me back to the beginning.

We are fortunate to be at the beginning of our AI for Business journey. I feel the thoughts of our professors and discussions with my cohort at a local Oxford pub, tugging at my pre-conceived beliefs – pulling them down and expanding my thinking of what is possible. Of course, we’ve got a lot of ground to cover together, but now I feel we have the best possible foundation to build upon for the next module together at Oxford.