I am AI Director at BT. I have a cross-cutting role in the Digital team to put in place the capability and capacity we need to address the amazing opportunities we have to harness data and AI to generate more value for our customers. There is already a lot of data science and AI going on in BT but we need to build upon that to accelerate our work and to spread the benefits and capability more widely within the company.
With millions of customers, not only in the UK but in the world, data and artificial intelligence play a pivotal role in providing the best experience to our customers. We sat down with Zoë Webster, Data and AI Director, to understand how we’re innovating in this space, some of the ethical implications and why people should consider joining us in this space.
Zoë, what do you do at BT?
How did you fall into Data & AI?
I have been fascinated by AI since reading my dad’s Isaac Asimov books as a youngster, both from the technological and philosophical side of things. Questions around the differences between human and machine intelligence led me, eventually, to study computer science and then specialisation in AI during an MSc including some study of philosophy of mind and cognitive psychology. I then joined an industrial R&D lab where I was able to develop and demonstrate cutting-edge AI approaches across several different applications. After that, and before joining BT, I spent 13 years at the UK Government’s innovation agency, Innovate UK, where I received a much broader education into the power of data and AI, as well as myriad other technology areas, across the economy.
How is BT innovating in the Data & AI space?
BT has a unique set of data sources as well as a unique place in the telecommunications industry. The complexity of our networks and the evolving needs of our customers requires us to innovate in data and AI. Currently, we are investing heavily but consciously into getting our data sources in the best shape so we can draw out the most valuable insights from them more quickly. We are also innovating in how we develop and deploy AI at scale, working with partners to accelerate this.
Personal data is something that more and more people are concerned about. What are the ethical implications of working in data and how do we mitigate risk?
When working in data, and particularly with personal data, we should all be mindful of potential ethical implications and not just when using automated processing. Combining two data sets for the first time, whether by a human or by a machine, could lead to new insights that might have implications on data subjects. Seemingly innocuous applications of data analysis could have serious consequences if not thought through carefully, particularly where personal data is used and when decisions may be taken based on the analysis. We mitigate that risk by raising questions about this explicitly to ensure those involved give risk due consideration. Guidance is provided to challenge assumptions and to steer mitigation. There should always be a human-in-the-loop who can make sense of what is being drawn out of the data. We are mindful of bias too, which stems largely from human bias, so raise questions about this in the guidance.
Why should someone consider a career in Data & AI at BT
Anyone working in AI wants to get their hands on real data and BT has real data in spades. Data about our customers and their interactions with our products and services, data about our network operations, data about our own processes and interactions between them all. If you want to work in data and AI, BT is the place to be. In addition, BT takes its responsibility very seriously and it is great to work across teams of people who care passionately about doing the right thing for our customers and colleagues. There are great opportunities here to learn in different roles and to learn and develop at the forefront of data and AI technology.
What kind of roles and skills are we hiring in this space?
We are hiring a large number of people across a wide range of skills, from data engineering and governance through to data scientists and AI developers, through to software, cloud and ML engineers and ethicists. In terms of skills, I would say that, alongside strong technical skills, the ability to collaborate well with other roles is a must.
Is BT doing anything to get more women into tech?
Aside from ensuring our job descriptions don’t use language that speaks to some people and not others based on their characteristics, BT is also working directly with sources of great female talent. An example of this is with CodeFirstGirls – we are working with their network to bring in a cohort of 15 individuals from the network to start work at BT.
You’re a senior leader in the Tech space. How important is to have more women and female senior leaders in this space?
I would suggest it is very important for two main reasons. We need to ensure AI deployments are not biased against any characteristics, resulting in discrimination, and this is less likely to happen if teams are inclusive. Discrimination based on sex has been seen in infamous AI experiments in settings where female AI developers, or any role of influence over the AI being developed, are in the minority. I believe that addressing this balance will ensure that we see less bias in AI. To do this, I think we need more women in AI to attract more women – we often look for people like us when deciding where to work. There are no easy fixes to this but we can start to make a difference. Already, I have heard comments from female interviewees that being interviewed by a women does make a difference to how they feel about a role and a company. We do need to watch though that women in AI are not asked to take on more than their fair share of the work here – that is where allyship comes in.
