The Trust Gap: Why 85% of Swiss Policyholders Expect Artificial Intelligence But Only 18% Welcome It

A recent Deloitte study reveals that while artificial intelligence has become a fixture in the daily lives of Swiss citizens, its use in the insurance industry is met with significant caution. Policyholders generally support AI applications that provide assistance, such as translating complex documents or offering safety tips, but they largely reject automated decision-making regarding coverage or premiums.

The research underscores a high demand for transparency, with a vast majority of participants insisting on human oversight for critical choices to ensure fairness. Although people are willing to share personal data for practical benefits like claims processing, they remain wary of how their sensitive information is utilized. Ultimately, the findings suggest that the insurance sector must establish clear ethical standards and robust monitoring to build consumer trust. This report serves as a strategic roadmap for firms to integrate emerging technology responsibly while addressing public concerns about algorithmic bias.

1. The Paradox of the Digital Daily

We are currently witnessing a profound cognitive dissonance in the Swiss financial landscape. Artificial intelligence has become a silent, high-frequency partner in our personal lives; yet the moment the algorithm touches a Swiss insurance policy, the consumer sentiment shifts from convenience to caution.

According to the April 2026 Deloitte study, "Insurance Firms and AI," 71 percent of the Swiss population now interact with AI at least once a month. More strikingly, 14 percent of citizens are daily users. However, this familiarity has not translated into institutional confidence. Only 18 percent of policyholders view AI in insurance as a positive development, while 36 percent remain overtly critical.

The industry is operating in a state of forced inevitability: 85 percent of respondents already assume that insurance firms are utilizing AI. For the strategic insurer, this data suggests that the battle for the customer is no longer about the technology itself—which is already assumed to be present—but about the transparency and agency afforded to the policyholder.

2. The "Decision Divide": Why We Love Assistance but Fear Automation

The data delineates a "Decision Divide" that serves as the new battleground for customer retention. Policyholders display a high degree of acceptance for "Supportive AI"—tools that enhance utility or simplify the user experience. However, there is a sharp rejection of "Decision-making AI" when it impacts financial standing or legal eligibility without human mediation.

This divide is best illustrated by the polar opposites of acceptance: 62 percent of respondents embrace AI for translating complex policy language, yet 61 percent flatly reject solely AI-based decisions regarding their applications.

High-Acceptance Use Cases (Supportive AI):

• Policy Translation: 62 percent accept AI simplifying complex policy jargon.
• Loss Prevention: 58 percent welcome AI-driven risk mitigation tips.
• Fraud Detection: 57 percent support AI in identifying and preventing insurance fraud.

Low-Acceptance Use Cases (Decision-making AI):

• Application Decisions: 61 percent reject AI making the final call on accepting or rejecting an application.
• Risk and Premiums: 41 percent oppose AI determining risk classifications and pricing.
• Behavioural Pricing: 37 percent reject the use of driving data to structure individual premiums.

As Madan Sathe, AI Lead for Insurance at Deloitte Switzerland, explains, "Our work with insurance firms has taught us that trusted AI needs clear rules and responsibilities. And the survey shows that customers expect transparency and monitoring. Specifically, this means that companies must design their AI systems in such a way that policyholders are informed transparently, humans make the important decisions, and regular checks are performed."

3. The Trust Deficit: A Crisis of Confidence in Oversight

The reluctance to embrace automated insurance is not a symptom of tech-phobia, but a targeted crisis of confidence in institutional oversight. Trust levels across the entire stakeholder ecosystem are jarringly low: only 14 percent of policyholders trust supervisory authorities to monitor AI, while tech providers (13 percent) and insurance firms (17 percent) fare little better.

This deficit is fuelled by a fear of "unfair decisions," a concern cited by 51 percent of respondents. This indicates that familiarity with AI does not breed trust when the stakes are high; rather, it highlights a perceived lack of algorithmic accountability.

Marcel Thom, Insurance Lead at Deloitte Switzerland, views this as a strategic signal, "The survey reveals a significant difference. AI has arrived in many people’s daily lives. However, attitudes are more varied in the insurance sector... This sends a clear signal to the industry as to how trust in the use of AI can be increased."

4. The Three Pillars of Policyholder Acceptance

For Swiss insurers to bridge this gap, they must pivot toward a human-in-the-loop architecture. An overwhelming 86 percent of respondents believe that important decisions should still be made by people. To meet this demand, the Deloitte study identifies three non-negotiable conditions that constitute the "secret sauce" for responsible AI deployment:

• Human Monitoring: 73 percent demand that a human check decisions in critical cases.
• Transparent Disclosure: 67 percent want clear communication when AI is utilized (rising to 85 percent in broader contexts).
• Simple Objection Processes: 59 percent want a streamlined method to challenge automated outcomes.

These pillars are not merely regulatory hurdles; they are the foundational requirements for consumer "agency" in an increasingly automated world.

5. Data Sharing: The Transparency Trade-off

The willingness of Swiss customers to share data is strictly transactional. While 56 percent are willing to share photos and documentation to expedite claims processing, this pragmatism vanishes when the purpose is opaque.

Willingness drops sharply in sensitive categories such as health metrics or driving data. Marcel Thom highlights a "transparency gap" where customers feel it is currently unclear how their data is actually being utilized. Without a clear "why" behind the data request, consumers will continue to opt for privacy over personalization.

6. Conclusion: The Human-Centric Future of InsurTech

The 2026 Deloitte study is not a rejection of innovation; it is a sophisticated guide to its survival. The path forward for the Swiss insurance market is not paved with more complex algorithms, but with more robust human-centric design.

The successful InsurTech leaders of the future will be those who move beyond the "black box" and facilitate a tripartite collaboration between regulators, tech providers, and policyholders to establish rigorous fairness standards. In this new era, the most valuable asset isn't the data—it's the trust of the person behind the policy.

Ultimately, we must ask: In an age of total automation, would you trust an algorithm with your most sensitive life decisions if there were no human safety net to catch it when it falls? For the vast majority of Swiss policyholders, the answer remains a firm "no."