How Contextual AI Boosts Underwriting Artificial Intelligence Adoption

Despite underwriting artificial intelligence (AI) being such a useful tool in the insurance industry, consumers are not fond of artificial intelligence when it comes to claims processing. Recent surveys found that only 17% of consumers are comfortable with insurance claims being reviewed solely by AI. In fact, 60% of consumers would rather switch companies than let AI review their claims. Why are people so distrustful of AI?

Unleashing the Potential of Contextual AI in Underwriting: Exploring the Benefits and Applications

This article explores that question, as well as deeper explanations of contextual AI and how it can be used in underwriting AI systems for better customer experiences.

What Is AI?

Artificial intelligence uses digital technologies to perform functions associated with human intelligence, such as learning from past experiences and predicting future outcomes. It encompasses technologies such as machine learning, natural language processing, and deep learning to build systems with human-like intellectual characteristics.

• Machine learning (ML): The technology uses algorithms to train computerized systems to learn from data without explicit programming. As machine learning programs ingest more data, they become more intelligent, delivering more accurate responses.
• Natural language processing (NLP): NLP helps AI understand human languages, whether spoken or written. Today's NLP uses reading and listening programs to translate language into constructs that computerized systems comprehend.
• Deep learning: Deep learning applies machine learning and artificial intelligence to create a hierarchy of learning similar to that of the human brain. For example, children learn the word cat by pointing to animals and saying cat. Adults respond with a yes or no. The brain collects these yes/no responses until the child can identify the characteristics that separate cats from other animals.

Although many AI implementations have been successful, there have been epic failures, such as Microsoft's AI bot spouting sexist and racist comments or Amazon's facial recognition flaws. It's these larger-than-life failures that lead people to question AI technology.

What Is Contextual AI (CAI)?

CAI uses AI algorithms to process information with the capability of adapting the acquired knowledge to new situations. For example, humans determine if the phrase "get out" is a command to leave or a statement of surprise by the information surrounding the statement.

Because people take their cue from tone, facial expressions, or body movements, they know if "get out" is expressing anger or surprise. Contextual AI takes into account these ancillary pieces of data before delivering a response.

How Is Contextual AI Used?

According to a Deloitte survey, 74% of respondents say they will be investing in artificial technologies in the upcoming year. At the same time, Deloitte points out that many companies are struggling to understand how to use the technology. Two key ways the insurance industry can use contextual AI are:

1. Underwriting Artificial Intelligence

McKinsey predicts that most insurers will have automated their underwriting processes by 2030. They will apply data analytics to internal data sources to assess risk and access external data for a more comprehensive evaluation. As convolutional technologies improve, underwriting artificial intelligence tools will be able to mimic human thought processes.

For example, suppose an applicant indicates that they do not participate in high-risk activities such as sky-diving or mountain climbing. However, contextual AI technology finds a thread on social media that the applicant is part of an indoor rock climbing club. Indoor rock climbing may not be as risky as mountain climbing. But there is a higher probability of an injury among those who rock climb versus people who do not. Flagging the activity is one way to ensure that premiums protect insurers against risk.

2. Customer Engagement

If insurers want to reach their customers and potential customers, they need to focus on social media. A recent survey found that cable TV usage dropped about 20% between 2015 and 2020. Plus, 56% of Generation Z and 66% of millennials say they are more influenced by advertisements on social media than streaming or video channels.

Given that the younger market is an insurer's likely target, companies need to look at technologies that can deliver customized ads in real time. This capability is especially helpful as the use of cookies declines. Contextual AI can deliver targeted messages that engage customers and keep them engaged.

What Are the Benefits of CAI as an Underwriting Artificial Intelligence Concept?

The public's awareness of unintentional biases in AI implementations is an obstacle to widespread acceptance of the technology in fields such as insurance. Contextual AI can help people become comfortable with AI because it is better able to explain how it arrived at solutions.

Minimizing the misinterpretations that have plagued AI can help build consumer and regulatory confidence. When the technology can explain the difference between shock and surprise in its interpretation of human interactions, for example, people can become more comfortable with the results.

1. More Data, Better Decisions

AI decisions are based on data; the more data sources, the better the decisions. Data repositories can hold external data collected through defined interfaces from multiple sources such as social media and online databases. CAI can then leverage that growing knowledge to make faster and more profitable decisions.

2. Better Transparency, More Trust

As more insurers turn to AI technology, transparency will become critical. Regulators will want to know more about how AI arrives at its conclusions. Today's AI systems often operate as "black boxes" where data goes in, the information is somehow processed, and outcomes are returned. The lack of transparency makes trusting the resulting decisions difficult, especially in highly regulated industries. Regulators will hesitate to approve AI models that cannot be explained.

Contextual AI adds explainability to the AI process. The technology can outline its processes in the context of human decision-making so that regulators and insurers understand the outcomes. This process enables developers to improve the systems to ensure fair and equitable decisions. Because of its contextual capabilities, AI will become an essential component of business operations.

3. More Efficiency, Better Profitability

With access to massive data volumes, contextual AI systems will continue to acquire new knowledge that can be applied to evolving market changes. The automated process will be able to analyze multiple data points in seconds to deliver a competitive quote to a potential client. This efficiency in highly competitive markets such as health and life insurance translates into higher profitability. 

Pilotbird offers an innovative contextual artificial intelligence technology that can move the insurance industry towards a fully automated value chain. It can minimize risk, engage customers, and reduce fraud. If you're ready to experience the power of contextual AI and underwriting artificial intelligence technologies, contact Pilotbird for a free demo.

Sources

Wiggers, Kyle. “Consumers Question AI-Driven Insurance Claims Review | VentureBeat.” VentureBeat, VentureBeat, 24 Aug. 2021, https://venturebeat.com/business/consumers-question-ai-driven-insurance-claims-review/.

“Reasoning.” Encyclopædia Britannica, Encyclopædia Britannica, https://www.britannica.com/technology/artificial-intelligence/Reasoning. Accessed 18 May 2022.

Burns. “What Is Machine Learning and Why Is It Important?” SearchEnterpriseAI, TechTarget, 30 Mar. 2021, https://www.techtarget.com/searchenterpriseai/definition/machine-learning-ML.

Lutkevich, Ben, and Burns. “What Is Natural Language Processing? An Introduction to NLP.” SearchEnterpriseAI, TechTarget, 2 Mar. 2021, https://www.techtarget.com/searchenterpriseai/definition/natural-language-processing-NLP.

Burns. “What Is Deep Learning and How Does It Work?” SearchEnterpriseAI, TechTarget, 29 Mar. 2021, https://www.techtarget.com/searchenterpriseai/definition/deep-learning-deep-neural-network.

Thomas, Anu. “Top 8 Funniest And Shocking AI Failures Of All Time.” Analytics India Magazine, https://www.facebook.com/AnalyticsIndiaMagazine/, 2 Mar. 2020, https://analyticsindiamag.com/top-8-funniest-and-shocking-ai-failures-of-all-time/.

Brdiczka, Oliver. “Contextual AI: The Next Frontier of Artificial Intelligence.” Adobe, 9 Apr. 2019, https://business.adobe.com/blog/perspectives/contextual-ai-the-next-frontier-of-artificial-intelligence.

Deloitte Insights. “2022 Insurance Industry Outlook.” Deloitte Center for Financial Services, Deloitte, 2022, https://www2.deloitte.com/content/dam/insights/articles/US164650_CFS-Insurance-industry-outlook/DI_Insurance-industry-outlook.pdf.

Balasubramanian, Ramnath, et al. “Insurance 2030--The Impact of AI on the Future of Insurance | McKinsey.” McKinsey & Company, McKinsey & Company, 12 Mar. 2021, https://www.mckinsey.com/industries/financial-services/our-insights/insurance-2030-the-impact-of-ai-on-the-future-of-insurance.

Inam, Rafia, et al. “Explainable AI – How Humans Can Trust AI.” Ericsson, https://www.ericsson.com/en/reports-and-papers/white-papers/explainable-ai--how-humans-can-trust-ai.