GAIBDS—GenAI Business Decision System—"Sense Early, Think Smart, Act Fast" Qorus-NTT DATA Innovation in Insurance Awards 2026

1.Concept and Objectives

As the insurance industry transitions from a growth phase into a mature phase, corporate strategy is shifting from pursuing scale to cultivating high-quality business, and operational approaches are evolving from "post-hoc analysis" to "forward-looking, precise value creation". The core goal of this project is to build an GenAI Business Decision-Making System, achieving "Sense Early, Think Smart, Act Fast"

Our objective is to reshape traditional business decision-making processes using GenAI and big data. Specifically:

- Sense Early (Intelligent Forecasting): Combine large language models, time-series models, regression models, and other advanced algorithms to build forecasting models that accurately predict future trends in core financial metrics such as premium, claims, and expenses.

- Think Smart (Intelligent Attribution): Based on the operational data of each policy, the root cause of business outcomes is identified through Chain-of-Thought (CoT) technology, providing a foundation for strategy formulation.

- Act Fast (Intelligent Decision-Making): Based on corporate objectives and attribution insights, use large language models to generate personalized, actionable operational strategies, enabling a shift from reactive responses to proactive anticipation—ultimately achieving faster and deeper operational insights.

2.Reasons Behind

The market has entered an era of saturated market, making extensive operating methods unsustainable. Traditional business strategies have long relied on manual judgment and subjective experience, leading to three core challenges:

-The outcomes are hard to predict. Manual forecasting struggles to process vast volumes of data, leading to significant deviations between predictions and actual results.

-Root causes are hard to identify. Identifying root causes relies heavily on manual efforts, which are slow and ineffective when dealing with massive datasets. This often leads to missed opportunities for timely strategic adjustments.

-Business decisions are hard to make. Decision-making lacks quantitative support and effect prediction, making it challenging to uncover latent business value.

Against these backdrop, there is an urgent need to adopt intelligent and refined operations to unlock new growth potential, address pain points in existing processes, and gain a competitive edge in the future market.

3. State of Competition

Within the industry, although some leading companies have begun applying traditional BI tools or basic machine learning for data analysis, the prevalent issue is that they focus primarily on presenting results rather than supporting actual decision-making. Most competitors remain at the stage of descriptive analytics (what happened), lacking the capability to leverage GenAI for diagnostic (why it happened) and predictive (what will happen) analytics. Existing solutions are largely auxiliary tools, unable to deliver end-to-end automated decision recommendations. By introducing GenAI to build an operational analytics bot, our project surpasses the current industry level in automated attribution and strategy generation.

4.Sources of Inspiration

The inspiration for this project comes from two main sources:

- Technology-Driven: The rapid advancement of GenAI, particularly large language models’ capabilities in natural language understanding, logical reasoning, and content generation, has inspired us to integrate "AI conversation" with "operational analysis."

- Demand-Driven: Due to operations managers' relentless pursuit of speed and accuracy, we extend process automation further into intelligent automation, aiming to create a 24/7 personalized intelligent business analytics bot that simulates the expert's thought process.

5.Departments Involved

This project is an innovative initiative involving cross-department collaboration, primarily engaging the following internal teams of Ping An Property & Casualty Insurance Company of China, Ltd.:

- Finance Department and Actuarial Department: This team provides business scenarios, historical data, and industry knowledge, and is responsible for validating the accuracy of model outputs and the feasibility of strategies.

- Architecture Team, Technology Center: This team is responsible for system integration and platform development, ensuring that AI capabilities can be seamlessly embedded into existing business systems.

- Data & Intelligence Platform Team, Technology Center: This team is in charge of GenAI model selection, training, fine-tuning, and the construction of computing infrastructure.

External partners involved include:

- Ping An Technology (Shenzhen) Co., Ltd.: Provides foundational AI platforms and big data technical support, enabling rapid integration of cutting-edge capabilities.

- Ping An Integrated Financial Services Co., Ltd. (Ping An Jinfu): Offers solutions and promotion strategies for data visualization dashboards and operational command centers.

6.Main Results So Far

Through the above efforts, the project has achieved significant outcomes:

- Industry-leading prediction accuracy: The forecast-to-actual deviation rate for business forecasts has been stably controlled within 1%, far exceeding the industry average.

- Exponential improvement in analysis efficiency: Attribution analysis time has been reduced from 1.5 days manually to just 10 seconds—improving efficiency by up to 12,000 times—with over 90% adoption rate of the analytical results.

- Support business decision-making: The system has generated 1,416 actionable strategic recommendations, effectively empowering operational decision-making.

- Breakthrough in business value: Since the project’s launch, user feedback has been overwhelmingly positive. Leveraging the platform’s capabiities, the company achieved its highest-ever profit in 2025, with direct value contribution from the platform amounting to tens of billions of RMB.

This project has successfully demonstrated the significant potential of GenAI in business operations, achieving a closed loop from data insights to business decisions-making.