Risk is defined by data and the more data you have, the more certainty you can afford in the insurance industry. So, insurers are hotly debating how data should be used, says Richard West, a practising solicitor and Global Head of Liability Defence as well as Global Head of Client Innovation at the UK-based Kennedys Law.

Kennedy’s has a presence in Technopark, Thiruvananthapuram, after its innovation division KennedysIQ merged with local company Cognitive Computing Services in 2018. The latter was set up by Tony Joseph, Jayakumar R and Renju VM in only the previous year. After merger, it was renamed Kennedy’s Kognitive Computing, the only development office outside of the UK.

Substantial legacy systems

Richard West explained to BusinessLine in an e-mail interaction how the insurance industry has been around for a long time. “Many of the leading insurers have been in business for many decades. They run substantial legacy systems, often different ones in different jurisdictions, across different lines of business.” A single insurer may have many hundreds of such systems across their portfolio.

Aggregating that data to inform risk pricing for efficient claims handling is difficult. “Insurers can be paralysed by collection of their data sets, deferring to underwriter’s intuition over data-informed insight. “Kennedys IQ and our lawyer colleagues are helping our clients to maximise potential of this valuable resource. It is better to prevent illness than treat it – so the same applies to all risk management,” he added.

Machines work alongside

Karim Derrick, Product and Innovation Director at Kennedys IQ, said as ability to use machines to predict and anticipate loss grows, so does the ability to prevent loss. “Explosion of driver aids in modern cars serves to substantially reduce death in vehicle-related accidents; traction control keeps us on the road; automatic breaking reduces shunts. So machines work alongside us to improve consistency of our driving.”

Commenting on the shift of insurance industry from ‘detect and repair’ to ‘predict and prevent’ mode, West said Kennedys IQ is particularly interested in how reputation is impacting modern businesses, increasingly based on the intangible value that includes reputation rather than physical assets like machinery and stock.

Assessing reputational loss

“We are working with underwriters to create new products that monitor and anticipate reputational loss by using machine to analyse social and traditional media in real-time, to machine-read contracts and agreements to ensure the right legal controls are in place across supply chains to ensure protection from environmental, social and governance risk,” he explained.

Reputation loss can be triggered by data loss, by misbehaving directors, poor treatment of employees or poor performance of products. Machines can be trained to recognise signs that these losses are occurring, helping insurers and corporates mitigate them in real-time in a way that the explosion of data and the power of machine intelligence affords.