Skip to main content

Generative AI in insurance

Does artificial intelligence have a place in the insurance industry?

It’s nearly impossible to go a day without hearing about the potential uses and implications of generative AI—and for good reason. Generative AI has the potential to not just repurpose or optimize existing data or processes, it can rapidly generate novel and creative outputs for just about any individual or business, regardless of technical know-how. It may come as no surprise that generative AI could have significant implications for the insurance industry.

Opportunities enabled by generative AI

Today, many enterprise organizations are finding opportunities to use generative AI in “horizontal” use cases enabling everything from dialogue generation for virtual assistants, to automated code generation, marketing and sales content generation, and much more. This convergence across industries allows organizations to leverage capabilities built by others to improve speed to market and/or become fast followers.

The insurance industry, on the other hand, presents unique sector-specific—and highly sustainable—value-creation opportunities, referred to as “vertical” use cases. These opportunities require deep domain knowledge, contextual understanding, expertise, and the potential need to fine-tune existing models or invest in building special purpose models. The real game changer for the insurance industry will likely be bringing disparate generative AI use cases together to build a holistic, seamless, end-to-end solution at scale. 

Unlocking value through investments in generative AI

Insurance organizations have a remarkable opportunity to create substantial value and realize the potential of generative AI by making well-thought-out investments that focus on three key value dimensions:

  1. Profitability and growth – Strategically directed investments can enable insurers to identify untapped growth opportunities, enhance product offerings, and expand market reach, ultimately driving profitability.
  2. Cost savings and efficiency – By investing in generative AI-driven solutions related to content creation for low-risk use cases, insurers can reduce spend across low-risk functional domains, thus enabling efficient spend allocation, which could lead to significant cost savings and operational efficiency gains.
  3. Operational intelligence and effectiveness – On an immediate basis, harnessing generative AI for autonomous coding is accelerating the software development life cycle resulting in productivity gains and reduction in training time, which may enhance workforce productivity.


Use cases for generative AI across insurance subsectors

Several generative AI use cases are gaining traction across insurance subsectors as insurers strive to strike the right balance between harnessing value and managing risk:

Property and Casualty (P&C)
  • Streamlined claims processing to enhance workforce productivity, enable cost savings and efficiencies
  • Loss prevention and control to enhance workforce productivity and generate new revenue streams
Life and Annuity (L&A)
  • Product personalization to generate new revenue streams
  • Agent assistance to enhance workforce productivity
  • Optimized underwriting and pricing to enhance workforce productivity and enable cost savings and efficiencies
Group
  • Customized group plans to generate new revenue streams and enhance workforce productivity
  • Improved member engagement to generate new revenue streams and enhance workforce productivity

Considerations to move forward

Potential risks and regulatory implications

Though the opportunities and value created by generative AI are impressive, artificial intelligence also introduces potential risks into the insurance industry. Insurance industry leaders would be wise to consider the following when scaling:

  • Malicious hallucinations and deep fakes, phishing and prompt injections, and ambivalent actors can expose the attack surface and erode customer trust.
  • Generative AI is prone to mimicking biases and propagating discriminatory behavior if implemented without guardrails and continuous monitoring.
  • Models will be trained on a corpus of proprietary and often private data, requiring regulatory compliance, node isolation, and source traceability.
  • Customer servicing and engagement within insurance companies requires a heightened sense of empathy and softer human interaction skills, especially during claim processing. Overemphasis on AI-driven automation may result in a lack of human touch, potentially leading to reduced customer satisfaction and loyalty.
  • Insurance regulators want oversight on insurers’ AI models and expect insurers to manage AI risk. AI oversight activity at the state level is forging ahead, with laws in place or contemplated, to bulletins from insurance commissioners asserting authority under multiple state and federal laws.
Mitigating challenges and moving forward

To minimize risk, insurance companies should prioritize the development of ethical artificial intelligence; leverage diverse and representative training data, evaluate, and audit their AI systems on a consistent basis through a robust governance model; and maintain transparency in decision-making. To learn next steps your insurance organization should take when considering generative AI, download the full report.

Generating possibility together

Driving business results with generative AI requires a well-considered strategy and close collaboration between cross-disciplinary teams. In addition, with a technology that is advancing as quickly as generative AI, insurance organizations should look for support and insight from partners, colleagues, and third-party organizations with experience in the generative AI space.

Did you find this useful?

Thanks for your feedback

If you would like to help improve Deloitte.com further, please complete a 3-minute survey