DigIn's virtual AI Summit 2023 takeaways

October 26, 2023 7:00 AM

As the realms of artificial intelligence (AI), machine learning and data science continue to gain momentum, insurance companies are harnessing their capabilities to thrive in an increasingly competitive digital landscape. The emergence of insurtechs has increased pressure on traditional insurers to further explore AI solutions that not only improve risk management, but also drive revenue growth and streamline operations.

Digital Insurance held the virtual DIGIN AI Summit 2023 on Oct. 24 to explore critical AI insights from industry leaders and experts, providing real-world insurance use cases including customer retention and acquisition, improved customer service solutions and innovative product development.

Digital Insurance is also hosting its Women in Insurance Leadership conference on Nov. 14, 2023 in Chicago, IL, that will include keynotes, round-table discussions and networking opportunities with some of the insurance industry's top women leaders in digital transformation.

The AI Summit's sessions revealed that insurance is ripe for digital transformation, and that AI and machine learning technologies are steering the industry in a new era. As insurtech pressure is driving insurance companies to embrace AI for risk management and operational efficiency, carriers must learn to navigate the new opportunities, and challenges, that innovative technology adoption brings.

The panelists emphasized the importance of robust data strategy and ethical considerations in AI implementation. In addition, they explored the industry's progress in technology adoption, recognizing the vast potential for innovation, particularly in claims, underwriting and sales. According to the experts, ethical responsibility and technological advancements that streamline key processes are critical in shaping the industry's future. Below are some of the sessions' highlights and key takeaways.

Panel 1: Creating and executing a comprehensive data strategy

John Broadrick, director of property & casualty product development at Nationwide, provided the significant components of a comprehensive data strategy.

"I always want to be looking for new data sources and new uses for the data that I already have, and I want to be able to keep track of the data that I have and be mindful of the controls needed around that data so I can accomplish my goals. And while I'm focused on that, I also want to make sure that I'm not misusing that data or allowing it to be misused by others," said Broadrick.

Sarva Muthukrishnan, head of global claims data and analytics at AXA XL, shared how insurers should determine who to include on a data strategy execution team, both internally and externally, to ensure they are getting accurate data.

"Execution requires concentration of various factors that include bringing together a strong capable team to execute on the strategy," Muthukrishnan said. "Process and technology are equally important, as well, but for the team, I would first identify areas of ownership, knowledge and expertise that are important to retain within the team, internally, and ensure protection of intellectual property."

He also added, "Partnership with tech companies [is] very important. We need technology to establish solid infrastructure, tools and software that can help drive the business. That's a critical factor. And we should not forget the role that data providers play in the data strategy overall for the execution."

Cathy Lanning, managing director and practice lead of financial services at Salesforce, highlighted the benefits and importance of utilizing accurate data information to improve the customer experience.

"Putting the customer at the center, understanding their journey and their experiences to your organization and making sure it can be proactive, making sure that the relationship can be recognized at all points and that they have consistency is really critically successful," Lanning said.

"There's a huge focus on how to organize my data around the customer, the single golden record or the single source of truth, so that I can understand my relationship... across lines of business, over time across channels, and the customer can move through a marketing sales and service experience in the same day," added Lanning.

Panel 2: Building the right talent and technology infrastructure

Effects from the pandemic and the rise of insurtechs have presented many new opportunities for carriers in the technology space. This session explored the current state of technology adoption in the insurance industry, and which areas of insurance have seen the most success in digitization or automation.

The session's panelists shared that overall, while insurance has made progress in implementing new technologies, the industry still has plenty of opportunity for innovation.

Darien Acosta, chief AI officer of Cover Whale, shared, "Coming from a more tech background and coming into insurance sort of as an outsider, I'm actually surprised regarding how far back insurance as an industry is in terms of technological adoption."

Acosta believes that as an inherent risk-averse industry, insurance is cautious and sees technology adoption as risky.

He stated, "I don't think insurance is putting its best foot forward because it wants to see what other people do first, and it doesn't feel like it's a leader in the space."

Bethany Jansen, strategic technology program manager of American Family Insurance, shared that there are some areas within insurance that have seen more technology adoption; claims first and foremost, followed by sales and underwriting.

"We have solutions across all three [processes], but starting in claims is probably where we've been able to do the most, and taken [adoption] further. In sales and underwriting we've made progress, however, increasing AI as part of these solutions is still something we are working on," said Jansen. "We don't always utilize AI or at least we don't always fully automate some of the processes. We do some semi-automation on steps that help alleviate some bottlenecks in a process; so summarizing large text, extracting information from imagery, et cetera."

Because the industry has some ground to cover in the digital space, there are still many opportunities for innovation.

Emmi Kim, senior data scientist at Farmers Insurance, said, "I think that a lot of the developments in technology are very exciting. We can serve the customers better by personalizing the product and also knowing better about what they want. And we can also assess the risk better as the insurance company to process the claims in a more efficient way. The whole development and evolution [of insurance technology] is very exciting."

Panel 3: Ensuring the ethical use of AI and machine learning for underwriting

In this final session, panelists discussed how carriers can assess risk data used for machine learning underwriting functions, how firms can ensure fairness when deploying AI and machine learning and the risks of using such technologies in underwriting.

On how firms should assess risk data for machine learning, Jamie Warner, managing director of Plymouth Rock Assurance, said that the most significant factor to consider is the legality and regulations surrounding the data.

"So I think the first thing that's helpful is obvious, which is what's legal. But that's not actually usually that beneficial in today's environment, because the ethical and moral and also the general use of data hasn't really caught up with the laws yet and we don't always know the direction regulators are going to go…" explained Warner. "[Firms should be] taking concern around the regulatory environment, and what regulators are saying and bringing them into the conversation. Also, thinking about the background of the different data providers, so have they worked with insurers before like reinsurers versus startups who may not be aware of some of the regulatory complexity that exists in insurance."

David Vanalek, chief legal and compliance officer at Richmond National, added that organizations should actively assess their data used for AI and machine learning to best ensure fairness.

"There certainly is a potential bias that folks need to mitigate against…" said Vanalek, "I do think it is certainly something that most companies take very seriously in trying to build a proper data governance and AI governance framework, for instance, while making sure it's cross-functional across the organization, and identifying what types of AI tools are being used for those particular tool sets, conducting proper assessments, seeing what the outputs are from those particular assessments and auditing that too."

He added, "I think that a key component of any good governance structure is aligning the values of the organization with the AI governance model and the other, building the models and then basically testing against that for disparate impact analysis."

The AI Summit's sessions revealed that insurance is ripe for digital transformation, and that AI and machine learning technologies are steering the industry in a new era. As insurtech pressure is driving insurance companies to embrace AI for risk management and operational efficiency, carriers must learn to navigate the new opportunities, and challenges, that innovative technology adoption brings.

The panelists emphasized the importance of robust data strategy and ethical considerations in AI implementation. In addition, they explored the industry's progress in technology adoption, recognizing the vast potential for innovation, particularly in claims, underwriting and sales. According to the experts, ethical responsibility and technological advancements that streamline key processes are critical in shaping the industry's future.