Detailed Look: AI-Powered Insurance Sector is Shaping up in 2023

This News Covers

• What are the Key Developments in AI Powered Insurance Sector in 2023
• Working of Predictive Analytics for Underwriting Explained
• How it Work, Benefits and Challenges
• Which are the fundraises that happened in 2023 in Insurance sector?
• Details of fundraising by Insurance players in 2023
• Digital wallet for insurance Marble bags $4.2m

The period between May and September 2023 marked significant advancements in the integration of chatbots within the insurance industry. As the digital age progresses, insurance companies are increasingly leveraging AI-powered chatbots to enhance customer experience, streamline operations, and reduce costs.

The period between May and September 2023 has solidified the role of chatbots in the insurance industry. As they become more sophisticated and integral to operations, insurance companies will need to stay abreast of technological advancements, ensuring they harness the full potential of AI-powered chatbots while maintaining the human touch that customers value.

MarketsandMarkets welcomes these developments and we take a look at the details here.

What are the Key Developments in AI Powered Insurance Sector in 2023

1. Personalized Customer Interactions:
   • According to NAIC, chatbots have evolved to provide more personalized interactions, understanding customer preferences and histories to offer tailored advice and product recommendations.
2. Advanced Cognitive Capabilities:
   • IBM's Watson Assistant for insurance has showcased advanced cognitive capabilities, enabling chatbots to understand complex queries, process vast amounts of data, and deliver insights in real-time.
3. Multifunctional Chatbots:
   • Research from AIMultiple highlighted the multifunctionality of modern chatbots. Beyond answering queries, they assist in claims processing, policy recommendations, and even fraud detection.
4. Zurich's Experimentation with ChatGPT:
   • As reported by Insurance Times, Zurich has been experimenting with ChatGPT, a cutting-edge chatbot model. This represents a significant shift towards more conversational and human-like interactions.
5. Zuri AI's Impressive Performance:
   • Zuri AI chatbot recently managed to resolve 70% of insurance queries end-to-end, showcasing the potential of AI in handling a majority of customer interactions without human intervention.
6. Funding Boost for Innovative Solutions:
   • Spanish insurtech, Inari, secured US$5.2 million in funding, emphasizing the growing investor interest in innovative insurance technology solutions, including chatbots.

Future Implications:

We foresee that the advancements from May to September 2023 are just the tip of the iceberg. The continuous evolution of AI will lead to chatbots that can predict customer needs, proactively offer solutions, and seamlessly integrate with other digital platforms. Moreover, as data privacy concerns grow, future chatbots will likely prioritize secure and transparent data handling, ensuring customer trust.

Working of Predictive Analytics for Underwriting Explained

Underwriting is the backbone of the insurance industry. It involves assessing risks associated with an insurance applicant and determining the terms and pricing of the policy. Traditional underwriting methods rely heavily on historical data and manual processes. However, with the advent of advanced technologies like AI, the landscape is rapidly changing.

Predictive analytics leverages statistical algorithms and machine learning techniques to identify the likelihood of future outcomes based on historical data. In the context of underwriting, it means assessing the potential risk of an applicant by analyzing vast amounts of data and predicting future claims probabilities.

How It Works:

At the heart of any AI system lies data. The richness and diversity of this data determine the potential accuracy and applicability of the AI model. In the vast sea of data, not everything is relevant or structured. Transforming this raw data into a usable format is a meticulous task. This is where the art and science of data science converge. It's about translating raw data into meaningful variables that the model can understand.

This is where the actual algorithmic magic happens. The chosen model, be it neural networks, decision trees, or any other, is trained using the processed data. No model is immediately perfect. Validation is the checkpoint to ensure the model's predictions align with real-world outcomes. Transitioning from a theoretical model to a practical tool, deployment is about integrating the AI model into the actual underwriting process.

The world is dynamic, and what was true yesterday might not be today. For an AI model to remain relevant, it must evolve.

1. Data Collection
   • Internal Sources: These are treasure troves of historical interactions with policyholders. Previous claims can provide insights into patterns of incidents, while policyholder details can give demographic and behavioral insights.
   
   
   • External Sources: These offer a broader context. Credit scores, for instance, can be indicative of a person's financial responsibility. Public records might reveal legal issues or past incidents, and social media activity can offer a glimpse into lifestyle and behavior, which can be particularly relevant for certain types of insurance.


2. Data Processing
   • Cleaning: This involves removing any inconsistencies, inaccuracies, or errors in the data.
   
   
   • Structuring: Data from different sources might come in various formats. Structuring ensures a standardized format, making it easier for algorithms to process.
   
   
   • Transformation: This involves converting data into a format or structure more appropriate for analysis, like normalization or categorization.


3. Feature Engineering:
   • Domain Knowledge: An intimate understanding of the insurance industry is crucial. It's not just about data but understanding what the data signifies.
   
   
   • Significant Predictors: Identifying which variables or combinations of variables can predict outcomes is a nuanced process. For auto insurance, while age and vehicle type are direct factors, combinations like age and driving history might provide more profound insights.


4. Model Development:
   • Generative AI's Role: Generative AI can create synthetic data samples. In situations where data is limited or imbalanced, these synthetic samples can provide a more holistic training environment, ensuring the model is robust and generalizable.


5. Validation:
   • Testing: Using a dataset separate from the training data ensures that the model's accuracy is genuine and not just memorizing the training data.


6. Deployment:
   • Real-time Assessment: As new applications arrive, the model instantly evaluates the data, providing risk assessments or recommendations, ensuring a swift and consistent underwriting process.


7. Continuous Learning:
   • Adaptation: As new claims data or policyholder interactions come in, the model adapts, refining its algorithms to reflect the most recent trends and insights.

The journey from raw data to an AI-powered underwriting tool is intricate, involving multiple stages of refinement, validation, and adaptation. Each step is crucial, ensuring that the final model is not just technically sound but also contextually relevant to the ever-evolving landscape of the insurance industry.

Benefits:

1. Efficiency: Automated risk assessment speeds up the underwriting process.
2. Accuracy: Predictive models can unearth complex non-linear relationships in the data that might be missed by human underwriters.
3. Consistency: Automated models ensure that similar applicants are assessed in the same manner, reducing biases.
4. Competitive Edge: Faster and more accurate underwriting can lead to better pricing strategies and customer satisfaction.

Challenges

1. Data Privacy: Handling personal data requires strict adherence to data protection regulations.
2. Model Transparency: "Black box" models can be hard to interpret, which can be a challenge in industries where explainability is crucial.
3. Over-reliance: Over-reliance on automated systems without human oversight can lead to errors or missed opportunities.

Predictive analytics is revolutionizing the underwriting process, making it more efficient, accurate, and data-driven. As a BU head, it's crucial to ensure that while we leverage the power of AI, we also maintain a balance with human expertise and judgment. The future of underwriting lies in this harmonious blend of technology and human insight.

Which are the fundraises that happened in 2023 in Insurance sector?

In 2023, the insurance sector witnessed significant financial momentum, with insurtech companies leading the charge in securing substantial investments. One of the standout fundraises was by bolttech, a Singapore-origin insurtech firm. On May 17, 2023, the company announced a whopping $196 million funding, elevating its valuation to an impressive $1.6 billion. Spearheaded by Japan's premier insurance entity, Tokio Marine, the funding round also saw participation from giants like MetLife and Malaysia’s sovereign wealth fund, Khazanah Nasional. With its innovative "embedded" insurance approach, bolttech has seamlessly integrated insurance offerings into consumers' purchasing experiences. Their expansive global network, encompassing over 700 distribution partners and 230 insurance providers, underscores their monumental growth since their inception in 2020.

Another notable fundraise in the same year was by Marble, a digital wallet specifically designed for insurance. On May 2, 2023, Marble secured a commendable $4.2 million in funding. This platform, which offers a holistic digital wallet experience for insurance premium payments, attracted investments from prominent names like Distributed Ventures, Blue Collective, and Goodwater Capital. With a vision to revolutionize the insurance experience, Marble is poised to address the challenges of 2023, emphasizing the importance of equipping users with robust tools to manage their insurance needs efficiently.

Details of fundraising by Insurance players in 2023

Insurtech bolttech gets $196M at $1.6B valuation from investors like MetLife

bolttech, an insurtech company that began in Singapore, has secured funding of $196 million, raising its valuation to $1.6 billion.

Date of Fund Raise: May 17, 2023

The funding round was spearheaded by Tokio Marine, Japan's premier insurance company. Other investors included MetLife through its subsidiary MetLife Next Gen Ventures and Malaysia’s sovereign wealth fund Khazanah Nasional. The company's Series B funding is noted as the largest straight equity Series B for an insurtech in the past year. Its Series A round in 2021 was also a record-breaker for insurtechs. The term "embedded" in insurtech refers to insurance or protection products integrated into the customer's buying experience. For instance, when buying a smartphone, a customer might be offered a protection plan.

Bolttech operates with a B2B2C model, linking over 700 global distribution partners with 230 insurance providers, offering 6,000 products to consumers. The company is recognized as a leading embedded insurance provider globally. Its clientele includes big names like Liberty Mutual, PayMaya, Progressive, Lazada, Samsung, and Home Credit. It operates across Asia, Europe, and all US states, quoting around $55 billion in annualized premiums. bolttech offers various embedded insurance products, with device protection being a standout. They collaborate with brands like Samsung, Windtre, LG U+, BackMarket, and Home Credit to offer these services. The company was established in Singapore in April 2020 and has since expanded internationally, thanks to partnerships with 700 distribution partners and over 230 insurers globally.

The funds from the Series B will be invested in bolttech's proprietary technology, emphasizing the incorporation of AI and machine learning in its operations. This includes computer vision, generative AI/natural language processing, advanced analytics, and robotic automation. There are also plans to enhance its insurance distribution technology, optimizing claims automation, fraud detection, and inventory management.

Digital wallet for insurance Marble bags $4.2m

Marble, a digital wallet and loyalty platform tailored for insurance, has secured $4.2 million in funding and introduced new features.

Date of Fund Raise: May 2, 2023

The funding round was spearheaded by Distributed Ventures (Chicago/New York). Other participants included new investors Blue Collective, Goodwater Capital, and CE Innovation Capital. Existing investors like IA Capital Group, MS&AD Ventures, Reciprocal Ventures, and several veteran angel investors from the finance and insurance sectors also took part.

Established in 2020, Marble serves as a comprehensive digital wallet that empowers users to shop, compare, explore, pay, and earn rewards on their insurance premium payments. Joining is free, and members can link various types of personal insurance, including healthcare, automotive, home, renters, pet, etc. In February 2021, Marble concluded its seed funding round, raising $2.5 million.

The company plans to utilize the latest funds to innovate new ways for its members to protect their loved ones and valuables. Specifically, Marble aims to grow its engineering and marketing teams, invest in in-depth user research, and develop unique features for the platform. Along with the funding announcement, Marble unveiled new features such as Rate Watch, Automated Shopping, and an Apple Wallet Integration. The Rate Watch feature notifies users about rate changes from major insurance companies that could affect their insurance premiums.

Stuart Winchester, CEO and founder of Marble, emphasized the company's mission to assist individuals in managing their policies and making informed decisions about renewals or shopping for new policies. He highlighted the challenges of 2023, with rising household expenses and insurance rate hikes, and stressed the importance of providing users with tools to manage their insurance effectively.