Insurance companies know that AI could transform their operations, but many struggle with outdated data systems and manual processes. A recent report highlights the biggest hurdles that prevent the industry from fully harnessing AI’s potential. Despite high optimism, actual AI adoption remains limited, mainly because of complex data issues and legacy systems.

Data Fragmentation and Legacy Systems Slow Down Progress

Many insurance firms are dealing with fragmented data sources that make it hard to implement AI effectively. On average, companies manage around 17 different data sources, which often become more complicated after mergers and acquisitions. This disjointed data landscape hampers efforts to establish clear governance and consistent data quality.

Legacy systems are another major obstacle. These outdated platforms are not designed to support modern AI technologies, making integration difficult. As a result, companies spend a lot of resources trying to reconcile data from different sources, which delays innovation and increases costs.

Operational Inefficiencies and Manual Errors Persist

Operational bottlenecks still drain resources. Many firms spend a significant portion of their budgets fixing manual errors—up to 14%. Reconciliation complexities also drive up costs, with about 22% of respondents citing this as a key issue. These manual tasks not only slow down processes but also increase the risk of mistakes and compliance risks.

Settlement cycles are often lengthy, with nearly half of companies operating cycles longer than 60 days. As transaction volumes are expected to grow by nearly 30% over the next two years, these inefficiencies could worsen if not addressed. Automating parts of these processes with AI could help, but current barriers prevent widespread adoption.

High Expectations vs. Slow AI Adoption

The industry is optimistic about AI’s future. About 82% of companies believe AI will dominate insurance operations eventually. However, only 14% have fully integrated AI into their workflows, and 6% have no AI use at all. This gap shows that while companies see AI as essential, many are held back by technical and organizational challenges.

One of the main issues is limited internal expertise. Many firms lack the skills needed to develop and deploy AI solutions. Additionally, integrating AI with existing systems requires overcoming technical hurdles, especially when data is scattered across multiple sources. This makes it hard to scale AI initiatives quickly and effectively.

The report suggests starting with areas like reconciliation processes, which are rules-based and easier to automate. Automating these boundary tasks could provide quick wins and demonstrate AI’s benefits. Ultimately, addressing data fragmentation and modernizing infrastructure are crucial steps for firms to realize AI’s full potential in insurance.

Inspired by

• https://www.artificialintelligence-news.com/news/for-effective-ai-insurance-needs-a-better-data-layer-and-integration/

Sources

• autorek.com