For Asia’s insurers, boldness is the key to an AI-driven transformation

There are numerous examples of Asian firms using AI to automate workflows or augment growth.

Insurance is about understanding risk and then using that knowledge to help people bounce back from distressing events. Artificial intelligence (AI) is likely to transform every step of this process.

That is a huge challenge to Asia’s insurance industry, and an equally massive opportunity. According to research, from 2020 to 2025, the insurance sector’s AI leaders created six times the total shareholder returns of the laggards. A key distinction of these leaders lies in how they rewire their organisations and reshape their people to build talent, culture, and organisational structures that allow AI to become fully embedded.

It is evident that the companies that are willing and able to do what a think tank calls “rewiring the enterprise” will see the greatest benefits.

Traditional AI excels at analysing and understanding data. Generative AI is building on that and has reached new levels of reasoning, judgment, and creativity. Agentic AI goes further still and could totally change interactions between insurers and consumers.  

Whilst Asian insurers are already using AI in sales, underwriting, claims, customer service, and back-office functions, few have turned that promise into bottom-line performance. The reason is that most of these efforts have been hit-or-miss, with lots of proofs of concept or isolated experiments. This can be useful, of course.

There are numerous examples of Asian firms using AI to automate workflows or augment growth. A Chinese insurer is deploying AI-powered visualisation capabilities to assess auto damages and accelerate the claims process; a Malaysian one is using gen AI to recommend products based on their customers’ browsing history.

But layering AI on top of existing processes is far from enough. Creating lasting value will only happen when insurers embed AI from top to bottom and across the organisation, with personalised service to customers at the centre of the action.

There are two main barriers to making an AI-driven transformation work: how insurers operate and, arguably, the harder of the two, how they engage and retrain human talent.

In terms of operations, a one-off use case is too small to affect profitability; transforming an entire domain, however, can make a sizeable difference: a 10 to 20% improvement in new-agent success rates and sales conversion rates, for example, or a 20 to 40% reduction in customer-onboarding costs.

Reaching this level of performance requires retooling workflows from beginning to end, including rethinking operating models and creating a modern data and tech stack. The goal: to find measurable improvement in unit economics. 

One major Chinese insurance group, for example, has developed a platform of multiple AI models that can take care of the entire sales process and provide substantial after-purchase customer service.

Because Asia’s insurers are competing in widely different economic and social contexts, the paths to AI transformation will vary. In almost every case, however, there are certain standard priorities.

First, the entire C-suite must be on board. And not just rhetorically. Start by focusing on a few important business areas, with defined metrics to evaluate outcomes. That way, success can be readily scaled.

Second, invest in flexibility. Building a modular tech stack, for example, in which AI components and capabilities can be reused, enables the organisation to evolve as the technology does.

Third, ensure that data is put to work. Insurers are repositories of an enormous variety of data.

But this has little value without the capabilities to use it. The ability to embed this data, including unique expertise and methods, into agentic AI systems could become core to insurers’ intellectual property.

Engaging people may be the more difficult challenge. Many insurers are on the same AI journey, so everyone is looking for talent, and there is not enough of it. At the same time, employees may fear AI and resist implementation. Successful rewiring requires energising the workforce so that people will work with AI in an integrated system. One promising approach is to orient teams around core “super products” across the value chain.

It’s important to emphasise that AI will not eliminate the need for human skills, far from it. But it is changing what people need to be good at. Organisations must therefore be prepared to provide relevant training, opportunities, and incentives. The ideal is for most talent-on the order of 70 to 80% to be in-house, so that innovations can spread widely and quickly.

Moreover, insurers must be clear that accountability cannot be outsourced to AI. People need to be in the loop and in positions to make the most important decisions. Insurance is a numbers business, rational and technical.

But it is based on human emotions and deals with people’s biggest fears. Trust is essential. Indeed, one important reason to invest in AI is to release leaders to concentrate on the primary care and creative thinking that consumers value, and that is central to the industry’s purpose.

None of this will be easy – this requires foremost true leadership to provide a strong change narrative, to role model, to set the right targets, incentives, and to support the capability building and change management at scale.  But companies that are willing and able to do this stand to gain the most.

The insurance industry has always changed with the times, but some times and some changes are bigger than others. Making the most of this moment will be difficult and expensive, requiring investment not only in legacy infrastructure and cybersecurity, but also in the people and organisational capabilities to make AI truly work.

Still, the costs of not starting an AI transformation are likely to be greater.