Discover the Hyper Challenge 2026 winners

Meet the winning team: CommPAEnion

This initiative provided valuable hands-on experience in how AI agents can collaborate across digital services to simplify insurance journeys. It also deepened our understanding of the standards, governance, and safeguards required to ensure agent-to-agent interactions are reliable, transparent, and customer- centric.​

Meet the winning team: Mythenqu-AI

We gained hands on experience in using AI to support smoother claims journeys, including more effective garage steering. We learned that timely, relevant repair options can help customers move forward faster, while transparent information and human oversight remain essential throughout the process.​

Meet the winning team: ZRoot

The initiative demonstrated how AI can help teams transform large volumes of information into actionable sales insights. The key challenge was separating signal from noise to provide underwriters with the most relevant data and recommendations. While the solution was built to address this specific use case, it was designed as a reusable capability, enabling the same architectural pattern to be applied across similar scenarios.​

Meet the winning team: CanAIda

We saw firsthand how AI can reduce manual effort and make complex underwriting information more accessible and actionable. The initiative also reinforced the importance of high-quality data and transparent insights, while recognizing that underwriting experts remain ultimately accountable for every decision. It further highlighted the value of maximizing organizational data assets and institutional knowledge to uncover cross-selling opportunities, provide a more holistic customer view, and support smarter, data-driven business decisions​

Meet the winning team: BBX

This initiative highlighted the value of making historical information easier to access and compare. It showed that AI can help reveal relevant patterns and support greater consistency, while underwriters retain responsibility for reviewing the information and making every final decision​

Meet the winning team: EH-I Canada

Beyond building a robust technical solution, we invested significant time with underwriters to understand their day- to-day challenges and identify where AI could deliver the greatest value. This user-centric approach enabled us to prioritize the features that mattered most, resulting in a solution that resonated strongly with stakeholders and supported underwriters’ decisions rather than replacing their judgement.​

Meet the winning team: Queens of the Danube

This initiative reinforced that personalized AI services must be built on a foundation of trust. Clear consent, transparent explanations, and robust safeguards against bias are essential to ensure customers remain informed, treated fairly, and in control of their decisions.​

Meet the winning team: Neural Nomads

Coming from different teams, we valued the opportunity the hackathon provided to collaborate on a new area of AI that fascinated us all. In addition to the automated research capability, we prioritized clear evidence trails and rollback mechanisms to ensure that outputs remained transparent, explainable, and easy to validate. This focus on trust and governance was a key differentiator of our solution. We named it Nomos: customizable, scalable, and guided by one principle - if you can score it, Nomos can improve it​.

Meet the winning team: Agentic Application Lifecycle

While much of the current focus is on coding agents, delivering successful solutions requires far more than writing code alone. We built a solution that leverages AI across the entire Software Development Lifecycle (SDLC), reducing friction for project teams and enabling them to focus on delivering greater value to their stakeholders.​

Meet the winning team: Token Boosters

Our team had limited prior experience in cybersecurity and none of us were weekend hackers. Despite this, we built a solution capable of rapidly identifying and exploiting vulnerabilities within the hackathon sandbox environment, demonstrating how AI can significantly lower the barrier to complex cybersecurity testing and analysis​.

Meet the winning team: Neon

Pricing work is resource intensive: reconciling data, stress testing models, interrogating portfolios, iterating. NEON's AI Pricing Agents work across the whole chain — cleaning and enriching data, exploring portfolios, developing pricing models, and rating individual risks.​

​ Actuaries and underwriters delegate to the Agents, who show their working at every step, so Zurich's experts stay focused where their experience matters: framing the right questions, judgement on key assumptions, and turning analysis into decisions the business can act on.​

Meet the winning team: TruthLens

We selected a deepfake-related use case because it is a growing concern for fraud investigators not only in Australia, but across the insurance industry worldwide. We focused on building a solution that combines probabilistic and deterministic techniques to optimize token consumption while maintaining high levels of accuracy. At the same time, we prioritized a robust audit trail, ensuring that expert reviewers could clearly understand, validate, and trust the solution's outputs.​