Insights from Huma Lodhi, Principal Data Scientist at Direct Line Group

As part of our Women in Tech series, we hear from Huma Lodhi, Principal Data Scientist for Artificial Intelligence and Machine Learning at Direct Line Group. Huma discusses her role, and makes some predictions for how AI and Machine Learning can help transform the insurance industry.

My career so far

I have now worked in tech for around 20 years, specifically in the fields of Data Science, Artificial Intelligence (AI) and Machine Learning. I spent a significant part of this time working in academia; I formerly worked as a researcher at a few Russel Group Universities including Imperial College London, University of Leeds and University of Warwick. I have also previously held the position of Senior Data Scientist, Manager at Cognizant, and ­worked for LV= before joining Direct Line Group.  I joined the Group in 2017, and my focus here is looking at how AI and Machine Learning can help transform our services.

I have a PhD in Machine Learning from University of London. During my career, I have co-edited two books: ‘Elements of Computational Systems Biology’; and ‘Chemoinformatics and Advanced Machine Learning Perspectives: Complex Computational Methods and Collaborative Techniques’, and I am proud to have had many research articles, with more than 2500 citations, published in leading AI & Machine Learning journals. I also hold the patent on an AI methodology: Support vector Inductive Logic Programming.

The role of AI and Machine Learning in insurance

I am absolutely fascinated by the role that AI and Machine Learning can play in the transformation of the insurance sector. I really believe that we are just at the start of the revolution, and I expect to see major developments in the field in the next five to ten years. 

The reason that there are so many opportunities for AI in insurance is because it is a data-rich industry, and this makes it attractive for data-driven Machine Learning techniques. These sophisticated methodologies can really transform insurance services ranging from risk management to claims settlement.  I think we will first see the benefits of advances in AI in claims processes. At present, the claims function requires a lot of human involvement, but utilising AI to replace a lot of what is done by humans will have many advantages. It will save money, which can be passed on to customers, and also enable us to provide better services. At the moment, it can take months to settle a claim and this can be really inconvenient for customers. If we replace these current claims systems with a new AI based system, it could help us to really speed that up.  Machine Learning techniques will also substantially improve risk identification and diversification that will lead to generation of fairer, accurate and personalized premium.

In addition, Machine Learning will help us tackle fraud. The insurance industry suffers billions of pounds worth of losses due to claims and application fraud and at present, it is quite hard to capture - fraud is a relatively rare occurrence, making it difficult to identify. However, Machine Learning algorithms have the theoretical foundation that makes them efficient at accurately detecting this activity and so it could have a huge impact in our efforts to fight it.

Common misconceptions and fears about Machine Learning and AI

A common fear about AI is that it will reduce jobs – I think people worry that they will become redundant as the machines take over. However, this is not the first industrial revolution that we have experienced. We have already seen the dramatic changes bought about through steam and electricity and, whenever there is a revolution, some jobs are lost and some are created. It is an opportunity for people to start to recognise the changes that are occurring, and to upskill accordingly. AI will actually lead to many more highly skilled jobs. So, whilst some jobs will inevitably be replaced by machines, the jobs that they create will balance this out. And I believe that it may offer us the chance to improve our work-life balance; to spend a bit more time with our families!

In summary, I believe that AI and Machine Learning have enormous benefits to offer, and as the technology develops and people start to understand it better, the fears will reduce.