Artificial revolution: Could AI genuinely change commercial insurance?

Authored by RSA

This is undeniably the year in which artificial intelligence exploded into public consciousness. We’ve seen numerous media headlines about how this rapidly evolving technology is about to change the world, prompting excitement and fear in equal measure.

A large part of the focus has been on generative AI, and many of us have both marvelled and chuckled at ChatGPT and Compose AI’s text production abilities and limitations.

Meanwhile, others have questioned the ethics of artificial general intelligence. The technology is still in its infancy, but could potentially have the power to surpass the human brain’s capabilities.

Yet hype aside, it’s undeniable that AI could have a significant role to play in commercial insurance in the years to come. In this blog we review some of the challenges and use cases facing the sector.

What threats does AI present to commercial insurers?

Arguably the biggest threat to commercial insurers as far as AI is concerned could come from within.

The problem, some commentators claim, is that the industry tends to be managed by an older demographic, which could stymie adoption.

Writing in Forbes, consultant Gene Marks advises that insurance businesses “need vision, self-confidence and the desire to adapt for the sake of growth and future profitability”.

But he warned that people who’ve “invested years in creating their own processes and systems” can be “reluctant to change” without a convincing return on investment.

At some point, though, commercial insurance business owners are going to have to embrace the new era, according to Marks.

That said, it’s essential not to confuse genuine concerns about AI with outdated attitudes.

Questions have been raised about algorithmic biases and the transparency of information sources, suggesting that there is some way to go in its evolution.

The way to ease these tensions, argues Rob Flynn, SME Director at RSA, is for the insurance sector to develop a collaborative approach to implementation.

"In this era of rapid change, it’s vital that important players across our industry, such as brokers, insurers, clients and other key stakeholders, join forces. This will enable us to identify the most effective and customer-beneficial ways of implementing AI. If we get it right, we will overcome the challenges and hurdles and maximise AI’s potential benefits for businesses and their customers." - Rob Flynn - SME Director at RSA

Can AI really add value in commercial insurance?

The answer to this question is most definitely yes, but with caveats.

The uses of AI for commercial insurers is growing.

For example, there’s the issue of underinsurance in the property market, thanks to the soaring cost of building materials in recent years. Some estimates claim as many as 79 per cent of commercial buildings are underinsured by up to 30 per cent (Fintech Finance News, 23 June 2023).

But AI platforms are now calling on a range of data sources to give building owners more accurate reinstatement values. This enables insurers to cover the full cost of claims, while adjusting premiums to accurately reflect the value of the property.

The caveat here is that although AI’s impressive data analysis can help solve the problem of underinsurance, human interaction will remain essential in commercial property. For instance, an insurance expert is able to spot unanticipated risks during a site visit that a remote computer cannot.

Another area in which artificial intelligence is proving transformative for the sector is fraud detection.

By being able to assess large volumes of data quickly, platforms can root out suspicious claims.

A fascinating case study here is French AI startup Shift Technology. Its fraud prevention services have processed tens of millions of claims with a 75 per cent accuracy rate in detecting fraudulent claims.

Artificial intelligence is also proving its worth in risk assessment, reducing the amount of time underwriters have to spend entering data, answering emails and editing documents.

Underwriters now have the ability to simply automate much of the information gathering component of the application process. The outcome can be faster quotes and lower costs.

It also frees up teams to work on other more valuable tasks such as policy tailoring and customer relations.

Should you adopt AI for your commercial insurance business?

Insurance is, and always will be, a people-to-people industry. This is particularly important in commercial insurance where claims can be complex and values vast.

Of course, artificial intelligence is showing how it can handle many tasks superbly and its capabilities are only going to improve.

But accountability is the watchword in this sector, and there will forever be a need for a human backstop when it comes to delivering commercial lines of business.