AI in Insurance: Streamlining Underwriting and Claims Processes

1. Introduction

The insurance industry, long considered a bastion of tradition and stability, is undergoing a profound transformation driven by artificial intelligence (AI) and machine learning technologies. This revolution is particularly evident in two critical areas of insurance operations: underwriting and claims processing. These processes, which form the backbone of insurance business models, are being streamlined, optimized, and in some cases, completely reimagined through the application of AI.

Underwriting, the process of evaluating risk to determine policy issuance and pricing, has traditionally been a time-consuming and labor-intensive task. It often involves the analysis of vast amounts of data from various sources, requiring significant human expertise and judgment. Similarly, claims processing, which involves assessing and settling insurance claims, has been a complex and often prolonged procedure, fraught with the potential for errors, fraud, and customer dissatisfaction.

The introduction of AI into these domains promises to address many of the challenges associated with traditional approaches. By leveraging advanced algorithms, big data analytics, and machine learning models, insurers are now able to process information faster, make more accurate predictions, and offer more personalized services to their customers. This not only improves operational efficiency but also enhances the overall customer experience, a critical factor in today's competitive insurance market.

This article aims to provide a comprehensive exploration of how AI is streamlining the underwriting and claims processes in the insurance industry. We will delve into the specific technologies being employed, examine real-world case studies and use cases, analyze key performance metrics, and discuss the broader implications of this technological shift. By the end of this piece, readers will have a thorough understanding of the current state of AI in insurance, its impacts, challenges, and future prospects.

As we embark on this exploration, it's important to note that the adoption of AI in insurance is not just a matter of technological advancement. It represents a fundamental shift in how risk is assessed, how customer interactions are managed, and how insurance products are developed and delivered. This transformation has far-reaching implications for insurers, policyholders, regulators, and the broader financial services ecosystem.

In the following sections, we will first provide an overview of AI in the insurance industry, setting the context for our more detailed discussions on underwriting and claims processing. We will then dive deep into each of these areas, examining the technologies, benefits, challenges, and real-world applications. Throughout the essay, we will support our analysis with relevant case studies, use cases, and performance metrics, providing a data-driven perspective on the impact of AI in insurance.

Let us begin our journey into the world of AI-powered insurance, where data-driven insights, automation, and intelligent decision-making are reshaping one of the world's oldest financial industries.

2. AI in Insurance: An Overview

Before delving into the specifics of how AI is transforming underwriting and claims processing, it's crucial to understand the broader context of AI adoption in the insurance industry. This overview will provide a foundation for our more detailed discussions in subsequent sections.

The insurance industry has always been data-driven, relying on statistical models and actuarial science to assess risk and price policies. However, the advent of AI and machine learning has dramatically expanded the industry's capabilities in data analysis, prediction, and decision-making. According to a report by McKinsey & Company, AI and its related technologies could potentially deliver up to $1.1 trillion in annual value to the insurance industry [1].

Key areas where AI is making significant impacts in insurance include:

1. Risk Assessment and Pricing: AI algorithms can analyze vast amounts of data from various sources to more accurately assess risk and determine appropriate pricing for insurance policies.
2. Customer Segmentation and Personalization: Machine learning models can identify patterns in customer data to create more refined customer segments and offer personalized products and services.
3. Fraud Detection: AI systems can detect patterns and anomalies indicative of fraudulent activities, helping insurers reduce losses and maintain the integrity of their operations.
4. Customer Service: Chatbots and virtual assistants powered by natural language processing (NLP) can handle customer queries and provide instant support, improving customer experience and reducing operational costs.
5. Predictive Maintenance: In property and casualty insurance, AI can predict when equipment or infrastructure is likely to fail, allowing for preventive maintenance and reducing claims.
6. Underwriting: AI streamlines the underwriting process by automating data collection and analysis, enabling faster and more accurate risk assessment.
7. Claims Processing: AI technologies can automate various stages of the claims process, from initial filing to damage assessment and settlement.

The adoption of AI in insurance is driven by several factors:

1. Increasing Data Availability: The proliferation of IoT devices, social media, and other digital platforms has created vast amounts of data that can be leveraged for more accurate risk assessment and personalized services.
2. Advancements in AI and Machine Learning: Breakthroughs in areas such as deep learning, natural language processing, and computer vision have expanded the capabilities of AI systems.
3. Changing Customer Expectations: In the digital age, customers expect faster, more personalized services, which AI can help deliver.
4. Competitive Pressures: As insurtech startups and tech giants enter the insurance market, traditional insurers are compelled to innovate to remain competitive.
5. Regulatory Changes: While regulations can sometimes be a barrier to AI adoption, some regulatory changes are encouraging the use of AI for better risk management and consumer protection.

According to a survey by Accenture, 85% of insurance executives plan to invest significantly in AI-related technologies over the next three years [2]. This high level of interest and investment underscores the transformative potential of AI in the insurance industry.

However, the adoption of AI in insurance is not without challenges. These include:

1. Data Quality and Availability: While there's more data available than ever before, ensuring its quality, relevance, and accessibility remains a challenge.
2. Regulatory Compliance: Insurers must navigate complex regulatory landscapes, particularly concerning data privacy and the use of AI in decision-making.
3. Ethical Considerations: The use of AI in insurance raises ethical questions, particularly around fairness, transparency, and potential bias in algorithmic decision-making.
4. Integration with Legacy Systems: Many insurers struggle with integrating AI technologies with their existing IT infrastructure.
5. Talent Gap: There's a shortage of professionals with the necessary skills to develop, implement, and manage AI systems in insurance contexts.

Despite these challenges, the potential benefits of AI in insurance are too significant to ignore. As we'll see in the following sections, nowhere is this more evident than in the areas of underwriting and claims processing, where AI is driving efficiency, accuracy, and customer satisfaction to new heights.

In the next section, we'll take a closer look at how AI is transforming the underwriting process, one of the most critical functions in the insurance industry.

3. AI in Underwriting

Underwriting is at the heart of the insurance business model. It's the process by which insurers evaluate risk, determine policy terms, and set premium prices. Traditionally, this has been a time-consuming, manual process heavily reliant on human expertise. However, AI is revolutionizing underwriting, making it faster, more accurate, and more efficient. In this section, we'll explore how AI is transforming the underwriting process, examining the technologies involved, the benefits realized, and real-world case studies.

3.1. Traditional Underwriting vs. AI-Powered Underwriting

To appreciate the impact of AI on underwriting, it's important to understand the traditional underwriting process and how it compares to AI-powered approaches.

Traditional Underwriting:

1. Data Collection: Underwriters gather information from various sources, including application forms, medical records, credit reports, and sometimes physical inspections.
2. Risk Assessment: Underwriters analyze the collected data to assess the level of risk associated with insuring the applicant.
3. Decision Making: Based on their analysis and experience, underwriters decide whether to offer coverage, and if so, under what terms and at what price.
4. Policy Issuance: If approved, the policy is issued to the applicant.

This process can take days or even weeks, especially for complex policies or high-value insurance.

AI-Powered Underwriting:

1. Automated Data Collection: AI systems can automatically gather and process data from multiple sources, including non-traditional data like social media and IoT devices.
2. Advanced Analytics: Machine learning algorithms analyze this vast amount of data to assess risk more comprehensively and accurately than traditional methods.
3. Real-time Decision Making: AI models can make underwriting decisions in real-time or near-real-time, dramatically reducing the time from application to policy issuance.
4. Continuous Learning: AI systems can continuously learn from new data and outcomes, refining their models and improving accuracy over time.

AI-powered underwriting can often complete the entire process in minutes or hours, even for complex policies.

3.2. Key AI Technologies in Underwriting

Several AI technologies are driving the transformation of underwriting:

1. Machine Learning: This is the core technology enabling AI-powered underwriting. Machine learning algorithms can analyze vast amounts of data to identify patterns and make predictions about risk.
2. Natural Language Processing (NLP): NLP allows AI systems to understand and extract relevant information from unstructured text data, such as medical reports or social media posts.
3. Computer Vision: This technology can analyze visual data, such as satellite imagery for property insurance or photos for auto insurance claims.
4. Predictive Analytics: By analyzing historical data, predictive analytics can forecast future risks and outcomes, helping insurers price policies more accurately.
5. Big Data Analytics: AI systems can process and analyze enormous volumes of data from various sources to gain deeper insights into risk factors.
6. Internet of Things (IoT): While not an AI technology per se, IoT devices provide valuable real-time data that AI systems can use for more accurate risk assessment.

3.3. Benefits of AI in Underwriting

The adoption of AI in underwriting offers numerous benefits:

1. Improved Accuracy: AI can analyze more data points and identify subtle patterns that human underwriters might miss, leading to more accurate risk assessments.
2. Increased Efficiency: AI-powered underwriting can process applications much faster than traditional methods, reducing turnaround times from weeks to minutes in some cases.
3. Cost Reduction: By automating much of the underwriting process, insurers can significantly reduce operational costs.
4. Consistency: AI systems apply the same criteria consistently across all applications, reducing human bias and errors.
5. Personalization: AI can enable more granular risk assessment, allowing for more personalized policies and pricing.
6. Fraud Detection: Machine learning algorithms can identify subtle patterns indicative of fraud that might be missed by human underwriters.
7. Improved Customer Experience: Faster processing times and more personalized policies lead to higher customer satisfaction.
8. Continuous Improvement: AI systems can learn from each underwriting decision and outcome, continually refining their models and improving performance over time.

3.4. Case Studies in AI-Powered Underwriting

Let's examine some real-world examples of how insurers are leveraging AI in underwriting:

Case Study 1: Ping An Insurance (China)

Ping An, one of China's largest insurers, has been at the forefront of AI adoption in insurance. They developed an AI-powered underwriting system called "AI Underwriter" that has transformed their operations [3].

• Technology Used: The system uses machine learning, NLP, and computer vision technologies.
• Implementation: AI Underwriter can process over 50 million insurance policies daily, analyzing text, images, and even video interviews with applicants.
• Results: Processing time reduced from 5 days to 10 minutes for some policies Underwriting accuracy improved by 90% Cost savings of over $750 million annually

Case Study 2: Lemonade Insurance (USA)

Lemonade, an insurtech company, has built its entire business model around AI-powered insurance, including underwriting [4].

• Technology Used: Machine learning algorithms and chatbots
• Implementation: Lemonade's AI, named "AI Jim," handles the entire underwriting process for home and renters insurance policies.
• Results: Policies can be issued in as little as 90 seconds Claims can be processed in 3 minutes Loss ratio improved by 40% compared to traditional insurers

Case Study 3: AXA XL's Ecosystem (Global)

AXA XL, a division of global insurance giant AXA, has developed an AI ecosystem for property underwriting [5].

• Technology Used: Machine learning, computer vision, and geospatial analytics
• Implementation: The system uses satellite imagery, weather data, and property information to assess risks for commercial properties.
• Results: 40% reduction in property survey costs Improved risk selection and pricing accuracy Enhanced ability to assess risks in remote or hazardous locations

Case Study 4: Tractable and Ageas (UK)

Ageas, a major UK insurer, partnered with AI company Tractable to use computer vision in motor insurance underwriting [6].

• Technology Used: Computer vision and machine learning
• Implementation: The system analyzes photos of vehicles to assess their condition and value at the point of underwriting.
• Results: Reduced need for physical inspections by 25% Improved accuracy in vehicle valuation Faster policy issuance times

3.5. Metrics and Performance Indicators

To quantify the impact of AI on underwriting, insurers track several key performance indicators (KPIs):

1. Underwriting Cycle Time: The time from application submission to policy issuance. AI typically reduces this from days or weeks to hours or minutes.
2. Loss Ratio: The ratio of claims paid to premiums earned. AI-powered underwriting often leads to improvements in loss ratios.
3. Expense Ratio: The cost of underwriting as a percentage of premiums. AI can significantly reduce this ratio through automation.
4. Straight-Through Processing (STP) Rate: The percentage of applications processed without human intervention. Higher STP rates indicate greater efficiency.
5. Underwriting Accuracy: Measured by comparing predicted risks to actual outcomes over time. AI often improves this metric.
6. Customer Satisfaction Scores: Faster, more personalized underwriting typically leads to higher customer satisfaction.
7. Fraud Detection Rate: The percentage of fraudulent applications detected during underwriting.

Here are some industry benchmarks and improvements seen with AI adoption:

• Underwriting cycle time reduction: 50-90% [7]
• Improvement in loss ratio: 3-5 percentage points [8
• Improvement in loss ratio: 3-5 percentage points [8]
• Reduction in underwriting expenses: 15-30% [9]
• Increase in STP rate: From 30-50% to 70-90% [10]
• Improvement in customer satisfaction scores: 10-25% [11]

These metrics demonstrate the significant impact AI is having on the underwriting process. As AI technologies continue to evolve and insurers become more adept at implementing them, we can expect to see even greater improvements in these areas.

4. AI in Claims Processing

Claims processing is another critical area in insurance where AI is making significant inroads. The claims process is often the moment of truth in the insurer-customer relationship, and its efficiency and accuracy can greatly impact customer satisfaction and the insurer's profitability. In this section, we'll explore how AI is transforming claims processing, the technologies involved, the benefits realized, and real-world case studies.

4.1. Traditional Claims Processing vs. AI-Powered Claims Processing

To appreciate the impact of AI on claims processing, let's compare the traditional approach with AI-powered methods:

Traditional Claims Processing:

1. Claim Submission: The policyholder submits a claim, often through a phone call or paper form.
2. Assignment: The claim is manually assigned to an adjuster.
3. Investigation: The adjuster investigates the claim, which may involve site visits, interviews, and document reviews.
4. Evaluation: The adjuster evaluates the claim based on the policy terms and the investigation findings.
5. Settlement: The claim is either approved, denied, or negotiated.
6. Payment: If approved, payment is processed.

This process can take days or weeks, depending on the complexity of the claim.

AI-Powered Claims Processing:

1. Automated Claim Intake: Claims can be submitted through digital channels and immediately processed by AI systems.
2. Intelligent Triage: AI algorithms automatically categorize and prioritize claims, routing complex cases to human adjusters.
3. Automated Investigation: AI can analyze photos, videos, and documents to assess damage and verify claim details.
4. Predictive Analytics: AI models can predict claim severity and potential for fraud.
5. Automated Settlement: For straightforward claims, AI can determine settlement amounts and initiate payments without human intervention.
6. Continuous Learning: The system learns from each claim, improving its accuracy over time.

AI-powered claims processing can often complete the entire process in hours or even minutes for simple claims.

4.2. Key AI Technologies in Claims Processing

Several AI technologies are driving the transformation of claims processing:

1. Machine Learning: This is the foundation of AI-powered claims processing, enabling systems to learn from historical claims data and improve decision-making over time.
2. Natural Language Processing (NLP): NLP allows AI systems to understand and process text-based information in claim forms, police reports, and other documents.
3. Computer Vision: This technology can analyze images and videos of damaged property or vehicles to assess the extent of damage and estimate repair costs.
4. Predictive Analytics: By analyzing historical claims data, predictive analytics can forecast claim outcomes, helping insurers manage reserves and detect potential fraud.
5. Robotic Process Automation (RPA): While not strictly AI, RPA is often used in conjunction with AI to automate routine tasks in the claims process.
6. Chatbots and Virtual Assistants: These AI-powered tools can handle customer inquiries about claim status and guide customers through the claims submission process.
7. Telematics and IoT: These technologies provide real-time data that can be used to verify claim details and even prevent losses.

4.3. Benefits of AI in Claims Processing

The adoption of AI in claims processing offers numerous benefits:

1. Faster Claims Settlement: AI can dramatically reduce the time from claim submission to settlement, improving customer satisfaction.
2. Increased Accuracy: AI systems can process vast amounts of data and apply consistent criteria, reducing errors and improving accuracy in claim assessments.
3. Cost Reduction: By automating many aspects of claims processing, insurers can significantly reduce operational costs.
4. Improved Fraud Detection: Machine learning algorithms can identify subtle patterns indicative of fraud that might be missed by human adjusters.
5. Enhanced Customer Experience: Faster processing times and the ability to submit claims digitally lead to higher customer satisfaction.
6. Better Resource Allocation: By automating routine claims, human adjusters can focus on complex cases that require their expertise.
7. Data-Driven Insights: AI systems can analyze claims data to identify trends and patterns, informing product development and risk management strategies.
8. Consistency: AI ensures consistent application of claims policies across all cases, reducing disparities in claim outcomes.

4.4. Case Studies in AI-Powered Claims Processing

Let's examine some real-world examples of how insurers are leveraging AI in claims processing:

Case Study 1: Lemonade Insurance (USA)

Lemonade, which we mentioned earlier for its AI-powered underwriting, also uses AI extensively in claims processing [12].

• Technology Used: Machine learning, chatbots, and computer vision
• Implementation: Lemonade's AI, "AI Jim," handles the entire claims process for simple cases, from initial report to payment.
• Results: 30% of claims are settled instantly Record for fastest claim processed: 3 seconds Claims expense ratio reduced to 2% (industry average is around 10-15%)

Case Study 2: Tokio Marine (Japan)

Tokio Marine, one of Japan's largest insurers, implemented AI in its auto insurance claims processing [13].

• Technology Used: Machine learning and computer vision
• Implementation: The system analyzes photos of vehicle damage to estimate repair costs and detect potential fraud.
• Results: 50% reduction in time required for damage assessment Improved accuracy in cost estimation Enhanced fraud detection capabilities

Case Study 3: Zurich Insurance (Switzerland)

Zurich Insurance implemented an AI-powered system to handle property and injury claims [14].

• Technology Used: Machine learning and natural language processing
• Implementation: The system reviews medical reports and assesses the severity of injuries to determine appropriate compensation.
• Results: Claims processing time reduced by 50% Consistency in claims decisions improved by 25% Significant reduction in operational costs

Case Study 4: Ant Financial (China)

Ant Financial, part of the Alibaba Group, uses AI for processing agricultural insurance claims [15].

• Technology Used: Machine learning, computer vision, and satellite imagery
• Implementation: The system uses satellite and drone imagery to assess crop damage and process claims automatically.
• Results: Claims processing time reduced from 7 days to 1 day 95% reduction in on-site visits required Significant improvement in farmer satisfaction

4.5. Metrics and Performance Indicators

To quantify the impact of AI on claims processing, insurers track several key performance indicators (KPIs):

1. Claims Processing Time: The time from claim submission to settlement. AI typically reduces this from days or weeks to hours or minutes for simple claims.
2. Claims Handling Expense: The cost of processing claims as a percentage of premiums. AI can significantly reduce this ratio through automation.
3. Customer Satisfaction Scores: Faster, more transparent claims processing typically leads to higher customer satisfaction.
4. Fraud Detection Rate: The percentage of fraudulent claims detected and prevented.
5. Straight-Through Processing (STP) Rate: The percentage of claims processed without human intervention.
6. Loss Adjustment Expense (LAE) Ratio: The cost of investigating and settling claims as a percentage of the claim amount.
7. Accuracy of Reserve Estimates: How closely the initial reserve estimate matches the final settled amount.

Here are some industry benchmarks and improvements seen with AI adoption:

• Reduction in claims processing time: 50-90% [16]
• Decrease in claims handling expenses: 25-30% [17]
• Improvement in customer satisfaction scores: 20-30% [18]
• Increase in fraud detection rate: 30-50% [19]
• Increase in STP rate: From 10-20% to 50-70% [20]
• Reduction in loss adjustment expense ratio: 3-5 percentage points [21]

These metrics demonstrate the significant impact AI is having on the claims processing. As AI technologies continue to evolve and insurers become more adept at implementing them, we can expect to see even greater improvements in these areas.

5. Challenges and Considerations

While the benefits of AI in insurance underwriting and claims processing are significant, there are also important challenges and considerations that insurers must address:

1. Data Quality and Availability: AI systems are only as good as the data they're trained on. Insurers must ensure they have access to high-quality, relevant data and that this data is properly cleaned and structured.
2. Regulatory Compliance: The use of AI in insurance is subject to increasing regulatory scrutiny. Insurers must ensure their AI systems comply with regulations regarding data privacy, fairness, and explainability.
3. Ethical Considerations: There are concerns about the potential for AI to perpetuate or exacerbate biases in insurance. Insurers must be vigilant in monitoring their AI systems for fairness and addressing any biases that emerge.
4. Explainability and Transparency: Many AI models, particularly deep learning models, operate as "black boxes," making it difficult to explain their decisions. This lack of transparency can be problematic in insurance, where decisions often need to be justified to regulators and customers.
5. Integration with Legacy Systems: Many insurers struggle with integrating AI technologies with their existing IT infrastructure, which may include outdated systems.
6. Talent Gap: There's a shortage of professionals with the necessary skills to develop, implement, and manage AI systems in insurance contexts.
7. Customer Acceptance: Some customers may be uncomfortable with AI making decisions about their insurance policies or claims. Insurers need to build trust and educate customers about the benefits of AI.
8. Cybersecurity Risks: As insurers increasingly rely on AI and digital technologies, they become more vulnerable to cyber attacks. Robust cybersecurity measures are essential.
9. Initial Investment Costs: Implementing AI systems can require significant upfront investment, which can be a barrier for smaller insurers.
10. Overreliance on AI: There's a risk that insurers may become overly dependent on AI, potentially overlooking important factors that AI systems might miss.

Addressing these challenges will be crucial for insurers to fully realize the benefits of AI in underwriting and claims processing.

6. Future Trends and Innovations

As AI technologies continue to evolve, we can expect to see several emerging trends and innovations in insurance underwriting and claims processing:

1. Explainable AI (XAI): As regulatory pressure for transparency increases, we'll likely see more emphasis on developing AI models that can explain their decisions in human-understandable terms.
2. Edge AI: Processing data at the edge (on devices rather than in the cloud) could enable faster, more secure AI-powered insurance services, particularly in areas like usage-based insurance.
3. Quantum Computing: While still in its early stages, quantum computing could potentially revolutionize risk modeling and complex calculations in insurance.
4. Blockchain and AI: The combination of blockchain and AI could enhance security, transparency, and efficiency in insurance processes, particularly in areas like parametric insurance and smart contracts.
5. Advanced Computer Vision: Improvements in computer vision technology could enable more accurate and automated assessment of property damage, medical imaging, and other visual data relevant to insurance.
6. Emotional AI: AI systems that can detect and respond to human emotions could enhance customer service in claims processing and personalize insurance products.
7. Predictive Maintenance: In property and casualty insurance, AI-powered predictive maintenance could help prevent losses before they occur.
8. Continuous Underwriting: Instead of periodic policy renewals, AI could enable continuous assessment of risk and dynamic pricing based on real-time data.
9. AI-Powered Risk Prevention: AI could increasingly be used not just to assess risk, but to actively help policyholders prevent losses.
10. Ecosystem Integration: AI could enable better integration of insurance services with other industries, creating more holistic risk management solutions.

7. Conclusion

The integration of AI into insurance underwriting and claims processing represents a significant shift in how the insurance industry operates. By leveraging advanced algorithms, big data analytics, and machine learning models, insurers are able to process information faster, make more accurate predictions, and offer more personalized services to their customers.

In underwriting, AI is enabling more accurate risk assessment, faster policy issuance, and more personalized pricing. The benefits include improved efficiency, reduced costs, and enhanced customer experience. Case studies from companies like Ping An, Lemonade, and AXA XL demonstrate the transformative potential of AI in underwriting.

In claims processing, AI is streamlining operations, reducing fraud, and dramatically improving the speed of claims settlement. The benefits include faster claims resolution, improved accuracy, and higher customer satisfaction. Examples from Lemonade, Tokio Marine, and Zurich Insurance showcase the significant improvements that can be achieved through AI-powered claims processing.

However, the adoption of AI in insurance is not without challenges. Insurers must navigate issues related to data quality, regulatory compliance, ethical considerations, and integration with existing systems. Addressing these challenges will be crucial for the continued successful implementation of AI in insurance.

Looking to the future, we can expect to see continued innovation in this space. Emerging technologies like explainable AI, edge computing, and quantum computing could further transform insurance operations. The focus is likely to shift increasingly towards predictive and preventive measures, with AI playing a key role in helping insurers and policyholders manage and mitigate risks.

In conclusion, AI is not just changing how insurance companies operate—it's reshaping the very nature of insurance. By enabling more accurate risk assessment, faster operations, and more personalized services, AI is helping to create an insurance industry that is more efficient, more responsive to customer needs, and better equipped to handle the complex risks of the 21st century. As these technologies continue to evolve, we can expect to see even more profound changes in the years to come.

8. References

[1] McKinsey & Company. (2018). Insurance 2030: The impact of AI on the future of insurance.

[2] Accenture. (2019). Technology Vision for Insurance 2019.

[3] Ping An. (2020). Annual Report 2019.

[4] Lemonade. (2021). Q4 2020 Shareholder Letter.

[5] AXA XL. (2019). AXA XL's North America Property Insurance business unveils Ecosystem of innovative technologies and services.

[6] Tractable. (2020). Ageas UK partners with Tractable to accelerate car insurance claims with AI.

[7] Deloitte. (2019). AI and the insurance industry: The long-awaited disruption.

[8] Boston Consulting Group. (2020). AI in Insurance: Hype or Reality?

[9] Capgemini. (2020). World Insurance Report 2020.

[10] EY. (2019). NextWave Insurance: large commercial and reinsurance.

[11] J.D. Power. (2020). 2020 U.S. Auto Insurance Study.

[12] Lemonade. (2021). Lemonade Claims Report 2020.

[13] Tokio Marine. (2019). Sustainability Report 2019.

[14] Zurich Insurance Group. (2020). Annual Report 2019.

[15] Ant Financial. (2019). Sustainability Report 2018.

[16] McKinsey & Company. (2019). Claims 2030: Dream or reality?

[17] PwC. (2020). Insurance 2020: The digital prize – Taking customer connection to a new level.

[18] KPMG. (2019). Insurtech 10: Trends for 2019.

[19] Coalition Against Insurance Fraud. (2020). The State of Insurance Fraud Technology.

[20] Willis Towers Watson. (2019). Quarterly InsurTech Briefing Q4 2018.

[21] S&P Global Market Intelligence. (2020). 2020 US P&C Insurance Market Report.