Is AI a Redefining Force Within the Insurance Industry?

Nigel Cannings is the founder and CEO of Intelligent Voice, the London-based company leading the international development of proactive compliance and technology solutions for various forms of media. It provides secure speech and NLP solutions to regulated and privacy-sensitive industries.

A qualified lawyer and legal expert, he has also worked for some of the world's largest law firms and software companies and is a recognised technology specialist and commentator. We caught up with him to find out more about how AI is transforming the insurance industry.

Q: How embedded is AI in the current insurance market?

The nature and core foundations of insurance companies remains unchanged, still built on risk with the focus on accurately assigning likelihoods of certain events. However AI has significantly changed the behaviour and demands of the consumer. Consumers are now digitally driven and more aware of the multiple product choices available to them but perhaps most importantly the patience and attention span of consumers has diminished in a world where we are used to getting results on the spot via omni channels. In order to respond to this market shift insurance companies have no choice but to embed AI technology. 

The effect of AI in the insurance industry was accelerated as a result of the Covid-19 pandemic and the focus on truly digital and automated processes whether that be for claims processing or underwriting, it is now at the forefront of most insurance company’s strategic plans and investments. AI is a prerequisite for some aspects of the automation and therefore has and increasingly will become a must-have component and business function for insurers to embrace and manage.  

Q: Why has AI become so central to the underwriting process? 

One of the biggest drivers behind the use of AI in the underwriting process is the sheer number of data sources required to process and ideally in real-time in order to calculate effective insurance pricing. The use of AI enables the number of data sources interrogated to be expanded thus providing a richer source from which accurate decisions can be derived with the goal of producing the most efficient return on investment for each premium underwritten.

The real-time processing also enables the customer experience to be enhanced as well as more business to be secured, in a world where a potential customer wants to know the price of their premium on the spot, AI can reduce the lifecycle to produce a premium from weeks to seconds. AI can also work to better customer experience by giving customer service agents less to think about so they can focus solely on the needs of the customer. Machine learning and voice recognition technology such as Intelligent Voice’s LexiQal, works behind the scenes whilst calls are taking place, managing to detect fraud at the earliest possible contact. 

Q: What trends in technology are we seeing emerge?

2022 will mainly see a continued focus on refining and implementing AI in the underwriting and claims arenas but some new trends are emerging. Usage based insurance is in demand given the awareness of waste identified through the pandemic and consumers are under increasing financial pressure through the rise in inflation and a rise in energy prices they want to pay for what they use type models. AI will be used increasingly in this domain as policy usage and behaviour will need to be tracked continuously and pricing updated in real-time. 

Another trend in 2022 will be the investment in AI security to proactively prevent fraudulent attacks on the insurance companies where 2021 was a record year for such behaviour. As we have found through our work at Intelligent Voice, AI can now be used to detect bad actors within the various insurance processes and triage the risk to the relevant fraud/security teams to investigate further. Voice recognition in Artificial Technology is now pushing the boundaries of fraud detection, including the analysis of audio and video data, which ensures that even in the increasingly technological world of customer interaction, fraud does not go unnoticed.

Q: Hyperautomation is emerging too. What differentiates it from ML, RPA and AI and why is it becoming popular?

Hyperautomation in the insurance sector is the next evolution in the use of AI in order to streamline operations, make smarter and more informed decisions and ultimately make insurance companies more profitable. Hyperautomation blends Robotic Process Automation (RPA) with AI and Machine Learning techniques to automate repetitive manual processes using data from multiple disparate sources.

The current automation achieved through RPA has its limitations and the complex data sets with non-linear workflows that are now part of the everyday insurance process mean insurance companies need to invest in blending these technologies together in order to support the next digital era of Hyperautomation.

Q: How has cutting-edge AI technology changed the culture of insurance companies over the past decade?

The biggest cultural change is the fact that insurance companies now have dedicated AI teams that solely work on these technology projects. These teams of individuals and skills which were not commonplace before AI became an embedded component in the insurance industry. The nature of these projects require access to multiple big data sources, access to expensive hardware for processing and the freedom to be creative, all of which are not natural occurrences within insurance companies. IT departments, security and compliance departments and engineering teams have had to adapt their culture and policies to enable the advances in AI. Change takes time and by its very nature creates friction but we are now seeing AI sit harmoniously within an insurance company’s setup and the culture. 

Q: How can AI help prevent the rising incidents of fraud?

There are so many different routes that an insurance fraudster can now exploit and with the increasing intelligence of these attacks it is key for insurers to invest in AI to reduce the risk. Insurance companies need to train the AI to have eyes across all of the various business processes and either flag the potential fraud or eliminate the ability for the fraudulent action to take place through proactive means.

The main routes being exploited currently are claims staging, bloating the value of claims, identity theft and ghost brokering. The risk on all of these as well as other routes can be significantly reduced through the use of voice recognition, AI and training large data sets which learn and develop over time. Voice holds more essential data than any other communication and voice recognition AI can recognise emotion, tone of voice, speech patterns and more to prevent fraudulent activity over the phone.

Also, it is of course the ability to process thousands of transactions per second rather than relying on manual reviews which makes fraud prevention and detection one of the biggest return on investment areas for insurers in the AI space. 

Q: The war on talent is a challenge for the insurance industry. Will this factor significantly slow down digital progress? 

The advancements in computer hardware has made AI accessible to the masses and also affordable, but the very nature of building, maintaining and enhancing AI models relies on access to specialist talent across researchers, engineers and data scientists. The most advanced AI models have the ability to automatically train and learn through self-learning loops but you still need engineers and data scientists available to maintain the code and infrastructure for which the models sit. Access to these talent pools is certainly an industry challenge being faced today. 

The two biggest talent factors affecting the speed at which AI can be adopted and deployed in the insurance industry is the global competition and demand for AI based talent across the majority of industries and secondly the lack of AI talent available who actually understand the intricacies of the insurance sector. Most of the AI talent in the insurance sector have to be trained and educated from scratch which of course slows down digital progress even further. Of course using AI in the talent acquisition process itself will of course improve an organisation’s chances of finding the right talent in the shortest possible time as well as partnering with academic institutions and developing within the organisation to nurture talent.  

Q: What does the future of insurtech look like to you? How do you see it developing?

The majority of start-ups that we see in the insurtech space have embraced the power of leading-edge technology and implemented the technology in processes and use cases which actually add value to the current insurance practices. It is more of a challenge for the larger well established insurance companies to embrace such technology at speed. We have seen ring fenced R&D divisions setup with insurance companies to improve the speed of implementation of new technology which is critical but this has had mixed success with traditional and sometimes bureaucratic legacy business processes often hindering such innovation. I believe we will continue to see insurtechs live side by side in partnership with insurance companies.

The global pandemic has accelerated the ability to transact pretty much anything online. This shift in market dynamics means it is vital to embrace technology and digital innovation and those that don’t will lag behind. It is the insurtech organisations that will act as the catalyst to enable this market shift.