Article on As demand for insurance and re...

Swiss Re has made artificial intelligence a core pillar of its Group strategy, with the reinsurer using AI to speed up policy issuance and improve underwriting capacity as demand for risk transfer and risk expertise grows, according to CEO Andreas Berger. The company reported net income of $2.8 billion for the first half of 2026, up 9% year-on-year. Speaking after the results, Berger said the risk landscape is becoming more complex and interconnected, which is increasing demand for the industry's services. "The risks are becoming much more interconnected and complex. As recurring supply shocks, geopolitical fragmentation, large-scale investments in new infrastructure reshape the global economy, the role of the insurance and reinsurance industry in helping businesses and societies build resilience is becoming increasingly important," said Berger.

AI speeds up policy issuance

Berger highlighted Swiss Re Corporate Solutions' international programme business as an example of where AI is changing operations. The industry average for issuing local policies is 45 days, he said. Swiss Re now issues them in under five days using data-driven rules, automation and AI-enabled workflows, with greater consistency and fewer handovers. The same approach is being applied to facultative reinsurance underwriting. Swiss Re receives more than 100,000 deal submissions each year in different formats from clients and brokers. "That volume is impossible to handle manually, and some opportunities have to be declined because they can't even be reviewed because of the sheer capacity and volume, and that is a constraint," Berger said. "By using AI, we can dramatically improve the triaging process and unlock capacity." AI now helps analyse each submission, identify the most attractive risks and flag issues that deserve attention. Underwriters can focus on the deals most likely to succeed, which supports faster decision-making and portfolio growth while maintaining underwriting discipline, said Berger.

AI deployment at scale

Swiss Re made AI tools like Copilot and ChatGPT available to all employees in 2024. More than 80% of staff now report confidence in using these tools, and around 14,000 of the firm's 15,000 colleagues work on a globally integrated data and technology platform that combines structured and unstructured data. "This provides the foundation for deploying AI consistently and at scale across the whole group, across all business units and group functions," said Berger. Berger said AI is deployed with clear human accountability, expert oversight and consistent standards, ensuring reliability for clients, regulators and shareholders. For executives looking at AI strategy, the case here is concrete: Swiss Re tied AI adoption directly to measurable operational outcomes, not experimentation. The takeaway for [AI for Executives & Strategy](https://completeaitraining.com/tag/executives-and-strategy/) is that AI value shows up in specific workflow bottlenecks - policy issuance times and submission triage - before it shows up in broad transformation narratives. The insurance applications are equally specific. [AI for Insurance](https://completeaitraining.com/tag/insurance/) professionals can point to Swiss Re's facultative underwriting example as a template: AI triages high-volume, varied-format submissions so human experts can focus on the deals that matter.

Why this matters for executives and strategy

Swiss Re's approach offers a working model for AI strategy at scale. The company identified two measurable pain points - 45-day policy issuance and 100,000 manual deal submissions - and applied AI to clear those bottlenecks. The results are quantifiable: policies issued in under five days, and underwriting capacity that no longer turns away business due to volume constraints. The broader point for executives is the sequencing. Swiss Re built a unified data platform first, then deployed AI tools across the workforce, then measured adoption and confidence. The strategy connects AI investment to specific revenue and efficiency outcomes, not vague productivity claims. That's the difference between AI as a pilot project and AI as a core pillar.