Emmanuèle Lutfalla and Louis Fer discuss how artificial intelligence is transforming the insurance industry, in Insurance Day

Introduction

Artificial intelligence (AI) is transforming industries at remarkable speed, and insurance is no exception. According to the Boston Consulting Group, it ranks as the second most advanced sector in AI integration, just behind technology, media and telecom. [1]

This is hardly surprising: insurers possess vast data reserves and analytical expertise, making the industry fertile ground for AI-driven solutions.

Although still at an early but accelerating stage, AI in insurance already shows immense potential to improve efficiency and innovation. Yet these benefits come with significant legal and ethical challenges that must be anticipated rather than ignored.

AI-Driven Efficiency and Innovation

The most visible use of AI in insurance remains customer-facing tools. Chatbots already handle 20–30% of policyholder requests without human involvement, cutting costs and freeing staff for more complex tasks. But the transformation goes much deeper.

Insurers now deploy AI across multiple operational areas. Advanced systems evaluate risks by analysing historical data and predicting potential claims, enabling underwriters to prioritize the most valuable or complex cases. Claims processing is accelerated through automated document analysis, image recognition, and pattern detection, reducing human error. AI is also widely used for fraud detection, identifying unusual patterns or inconsistencies that may escape traditional review methods. Beyond operational efficiency, AI provides insights for product development, helping insurers design coverage tailored to emerging customer needs.

AI-driven pricing is another transformative element. By integrating real-time data from connected devices, usage-based policies, and external risk indicators, insurers can adjust premiums dynamically, ensuring fairer, more precise pricing.

This not only benefits insurers but also offers customers more personalised products and transparent pricing structures.

Overall, AI enables insurers to handle routine tasks efficiently, focus human expertise on complex decisions, and deliver a more responsive and customer-centric service model.

Legal Risks and AI-Created Challenges

With every technological advance comes new legal and regulatory challenges. This is especially visible in Europe, where the EU AI Act (adopted in 2024) establishes a framework to ensure that AI is trustworthy, human-centric, and consistent with fundamental rights.

The Act divides AI systems into risk categories. High-risk systems, such as those used in health or life insurance underwriting, face strict requirements on data quality, fairness, transparency, documentation, human oversight, and registration with supervisory authorities.

Limited-risk systems, such as customer-service chatbots, face lighter obligations, while minimal-risk systems remain largely unregulated.

The AI Act has particular implications within the insurance system. Pricing and underwriting in health and life insurance fall under the “high-risk” category, requiring compliance with strict rules on data integrity, model accuracy, and transparency. Providers must also maintain detailed technical documentation and allow effective human oversight. Limited-risk applications, by contrast, such as customer-service chatbots or tools used in non-life insurance, including property and liability coverage, require only clear disclosure and user opt-out mechanisms.

Penalties are severe: fines can range from 1% to 7% of global annual turnover or EUR 7.5–35 million. For insurers, compliance failures are not minor infractions but existential threats to reputation and revenue.

Building Trustworthy AI: Governance, Compliance and Legal Support

As AI becomes more embedded in insurance operations, effective governance is essential.

Beyond technical safeguards, insurers must put in place clear structures to oversee their systems from design to deployment and monitoring.

• Mapping AI systems: taking stock of all tools, whether developed in-house or externally sourced, and classifying them under the relevant risk categories.
• Embedding lifecycle controls: requiring conformity assessments, thorough documentation, and regular audits of models; defining clear rules for training data, human review, and incident response.
• Integrating legal expertise: involving lawyers familiar with both technology and insurance law helps ensure compliance, reduce liability, and align deployment with evolving regulations.

Legal guidance is particularly valuable when drafting contracts with technology providers, reviewing compliance with applicable regulations and preparing regulatory filings, or assessing liability in case of disputes. A proactive legal strategy not only prevents penalties but also enables insurers to take full advantage of AI’s potential without undermining trust.

Conclusion – Balancing Innovation and Responsibility

AI offers insurers unprecedented opportunities to streamline operations and deliver more personalized products. Yet the same tools that drive efficiency can also create regulatory and ethical risks if deployed without proper oversight.

Insurers who treat governance and legal strategy as integral to their AI efforts, rather than as afterthoughts, will be best positioned to succeed. Success in the age of AI will depend on maintaining a balance between innovation and responsible practice.

[1]: Insurance Leads in AI Adoption. Now It’s Time to Scale. Joe Khoury, Daniel Martines, Christopher Freese, Jürgen Eckel, Malli Gupta, and Lara Najjar, 4 September 2025.