Welcome to the Insurance AI Pioneers Awards 2018

Insurance Nexus AI Pioneers Awards 2018

At the Insurance AI & Analytics USA Summit 2018, we will be exploring where and how carriers are taking advantage of AI and machine learning to drive business impact. In 2018, Insurance Nexus would like to recognize the leaders and innovators that are daring to risk and lead the path on AI-Driven transformation.

We now have our nominees, so we are asking the Insurance Nexus and wider community to vote on the winners. Awards will be announced live onsite at the Insurance AI and Analytics USA Summit, during the Networking Drinks function in the evening of Wednesday, June 27th.

Question Title

* 1. Underwriting Application: this award recognizes the most exceptional business case on underwriting over the past 12 months

TruMatch, by Marsh

A major objective of insurance brokers is to match their clients’ risks with the right carriers. Although this sounds easy, it can be quite challenging. Brokers are not always aware of every carrier’s latest underwriting preference and conducting an exhaustive market search can be impractical and time consuming.

Commercial underwriters typically decline 30% to 90% of the submissions they receive because they do not match their risk or pricing appetites. Not only is this inefficient, but in some cases, it leads to an incomplete list of all available carriers for a client’s particular risk.

To solve for this, Marsh’s Insurer Consulting Group launched in 2017 TruMatch, a multi-platform accessible application powered by Artificial Intelligence that combines the machine learning principles used in recommender systems available in e-commerce and social media with Marsh's transactional database to provide carrier and prospect suggestions in two substantive ways: Identifying suitable carriers for clients

For Marsh brokers, TruMatch enhances their ability to identify suitable carriers for a client’s specific risk. Brokers select the account or prospect characteristics and in just a few seconds TruMatch evaluates and returns potential carriers and relevant marketing collateral. Carriers are ranked based on factors including past quoting activity and carrier performance ratios.

For carriers, the TruMatch’s algorithms analyze Marsh’s data to identify potential clients that best match their historical underwriting preferences.

As the criteria used to evaluate a risk may differ for every product, TruMatch considers the importance of the variables inputted and incorporates strategic and product considerations into the matching process.

Tensorflight

Tensorflight’s vision is straight through processing of commercial property insurance. They are using satellite, aerial, drone and street view imagery to automatically inspect property from each angle. They can return the majority of commercial property underwriting factors (e.g. square footage, construction type) in seconds for any building in the USA.

Before Tensorflight, AI insurance companies had to wait weeks for loss control engineers to inspect property and paid hundreds of dollars for a single property inspection. Now, Tensorflight’s AI can do the same inspection in seconds at a cost of a few dollars.

Planck Re

Planck Re combines Artificial Intelligence and open-web data-mining to provide commercial insurance companies with a data platform that substantially enhances the quality and speed of the underwriting process, reduces onboarding friction and streamlines renewals.

The company is working with several top tier insurance companies in the U.S & world-wide. For example, in the past few months they’ve been working with a top tier insurance company in the states, providing them technology through API that was integrated with their agents' quoting portal, to automatically answer the underwriting questions, after submitting only the business name & address.

The pilot was tested on 7 SMB segments and resulted a coverage of 99.1% of the underwriting questions, all with over 90% accuracy. For example, one of the segments was Photographers & Videographers, in which one of the underwriting questions was whether they use aerial imaging equipment.

To answer this question after receiving only the name & address via the API, their platform automatically identified images or videos taken by the relevant photographer and using specially trained deep learning models analyzed and concluded whether it was taken from the air or not. This application is now being expanded to additional segments, answering many other types of questions.

Cape Analytics

Today, on average, 30 percent of insurer property data around critical attributes is inaccurate. This is a massive problem for insurers that affects policy quotes and underwriting. Cape Analytics closes that gap significantly by providing more accurate property risk information -- such as geolocation, roof type, condition, building footprint, and overhanging trees – instantly, at time of quote.

In order to extract these key property attributes, Cape Analytics use cutting edge technology, including geospatial imagery, advanced computer vision algorithms, and data science. Cape Analytics leverages cloud compute to do this at massive scale, extracting data for over a million homes per hour across the United States. Compare this to the data sources insurers relied on previously: tax assessor data can be years, if not decades old, while in-person inspectors are expensive to deploy, and data is not available at time of quote. Cape Analytics provides a data source that is instant, accurate, and easily scalable.

Cape Analytics has seen impressive growth in the last 12 months, expanding from five million residential buildings in Florida, to a nationwide offering with data on 70+ million buildings, and signing a partnership with Duck Creek. During that time, we have also added regional, super-regional, and national insurers and reinsurers as customers, including Security First Insurance, Nephila, XL Catlin, as well as many others. Meanwhile, early customers are re-upping and doubling down on their contracts with Cape Analytics. The combination of a nationwide footprint, as well as a proven track record of product and customer success has led to enormous momentum for the company as a leading underwriting solution.

Pod, by Hippo

Current Policy Management Systems (PMS) are cumbersome, lack agility and require months, if not years of IT resources to make changes and launch new programs. These limitations are the reason Hippo decided to build their own system from scratch. They designed a cloud-based PMS lovingly nicknamed Pod (a group of Hippos) which rolled out in April 2017. Pod is a cutting-edge PMS solution that is accessible to partners, agents & underwriters whenever and wherever they're connected online.

The Pod system offers all the functionality that underwriters and agents need within a single system including billing, document generation and storage, quoting, binding, underwriting, and servicing. It creates a streamlined on-boarding process and a superior underwriting experience through:

- Unlimited expansion and storage, enhanced data encryption, redundancy backups, and access from anywhere in the world
- Automated policy documents that are generated in real-time once a transaction is completed
- Cancellations, non-renewals, non-payments and reinstatement notices are automated to be sent to the customer. Notices can be sent via email, in real-time, allowing enhanced transparency and speed, maximizing customer response time - and backed up with a mailed letter.
- Automated endorsement processing, fully prepared for future effective dates, allowing underwriters to pre-program future endorsements or cancellation updates without any follow-up processing required

AIG

Clients and brokers had asked AIG improved clarity on how cyber risks are underwritten, specifically on how their risk is viewed and how their cyber-security risk posture compares to peers’; a way to better differentiate their risk from peers; and recognition for what they are doing well. Last fall, they announced their solutions: a new cyber insurance application, underwriting model, and CyberMaticsSM.

AIG now provides greater transparency on how they underwrite cyber risks by providing scoring and analytics reports to help applicants better understand their areas of strength and weakness. Even applicants who do not bind coverage can receive a basic report; while those that do bind receive more detailed information, including cyber maturity scores, baseline risk trending, top risk scenarios and risk reducing controls, and more.

Insureds’ CISOs can use the information in the reports to evaluate their organization’s threat profile and resilience in detail and further use the underwriting model’s mapping on how different controls affect each risk scenario to adjust their cybersecurity. The reports are driven by AIG’s new cyber application that is dynamic in that it adjusts the questions asked based on the applicant and coverage sought.

Furthermore, CyberMatics, AIG’s patent-pending security services-led approach, helps them conduct an even more accurate, ongoing risk assessment of an insured’s IT environment. CyberMatics provides validated answers to the application questions and gives insureds the ability to access an up-to-date risk assessment of their technology control and threat environment. Clients engage with Crowdstrike or Darktrace to regularly provide translated data relevant to the application, allowing for greater accuracy and improved scoring with the potential for improved terms depending on their cyber maturity. Clients can rest assured that their raw data is translated before coming in, and only data directly pertaining to the underwriting application questions is needed.

Question Title

* 2. Customer Experience Application: this award recognizes the most exceptional business case on customer experience over the past 12 months

Intelligent Water Monitoring Solution, by Eddy Home

Water damage has surpassed all other kinds of property perils to become the number one risk, accounting for billions in damages annually in North America. Water damage can be immediately mitigated with Eddy’s Intelligent Water Monitoring Solution, in both single dwelling homes and multi-family/commercial buildings (where water damage risk is multiplied).

Their IoT-powered technology enables leak detection and preventative measures for owners, property managers, and building operators, resulting in the proactive detection of issues and subsequent water savings/resource management. Their suite of products offers the ability to monitor and control water usage, protecting the property and owners/tenants against the risk of water leaks and damage. Eddy’s suite of products includes a monitoring solution that tracks all usage in the property with a remote and automatic shutoff. The result is a significant reduction in water damage claims.

Safelite Alexa Skill and Watch Us On Our Way, by Safelite

Safelite® AutoGlass enhanced the customer experience by introducing a Safelite Alexa skill and an improvement to 'On My Way Text', called 'Watch Us On Our Way'.

The Alexa Skill offers voice-control solutions leveraging a new environment for the brand. Developed by Safelite Works (Safelite’s innovation lab) the Alexa skill went from proof of concept to live within a few weeks of time, offering consumers several ways to get their auto glass repaired. Users can say things like: “Alexa, tell Safelite I cracked my windshield,” or “Ask Safelite where the nearest location is,” or “Tell Safelite to get a quote,” and more.

The Alexa Skill is a success. It demonstrates best practices in voice control, and a better understanding of effective ways to engage a consumer in audio-based interfaces. The Skill includes cross-device integrations allowing it to prompt a text message, directing the user to Safelite’s website for data input to schedule an appointment.

Watch Us On Our Way capitalizes on technology similar to Uber or Lyft. Customers see the exact location of a technician en route to their location. Customers can elect to receive a text when their technician is on the way—one supplemented with a link for a map view that’s updated in real-time including their technician’s location. The customer knows exactly when the technician will arrive so they can prepare.

WEX Dashboard, by DXC Technology

DXC first implemented a WEX based Dashboard to provide a unified comprehensive view of customer and policy information. With it, phone agents now quickly access policy details from a centralized desktop. To fully realize long term operational efficiencies, DXC identified an additional opportunity to reduce the amount of time agents spent looking for policy information and department procedures. DXC partnered with IBM Watson to implement a solution providing phone agents quick and accurate answers, in turn reducing or eliminating the need, to contact an internal assist line.

The deployed solution, a virtual assistant developed using Watson Conversation (WCS) and Discovery (WDS) services was integrated into the WEX Dashboard. The virtual assistant provides agents the ability to search for answers during or after a call. Agents are also able to conduct a quick search of unstructured information for answers that shortens the research time to find information. Within 60 days of launch, the population demonstrated a reduction in the number of calls to their internal assist line by over 30%.

A key indicator of the success of this launch is the quantifiable improvements in the customer experience. Phone agents are much more confident and better equipped to quickly respond to customer inquiries. Quality has improved as phone agents are providing complete and accurate data to the caller. Call handle times continue to decrease as phone agents discontinue usage of the assist line and become experts at navigating the deployed solution. Collaterally, first call response has also improved as customers are more confident in the information received during the first call and inquiries do not require multiple calls to resolve. These factors are also reducing overall operating expenses. This is a storyline of long term sustainability and a true transformation of a customer team with AI becoming a competency.

Life.io

Life.io powers policyholder experiences for the leading insurance companies, including 2 top 5 carriers.

Their data science and predictive analytics team has enabled them to enhance the platform and the policyholder experience for clients in the following ways:

1. Hyper-targeted and personalized experience: The content and experience within the platform adapts and is personalized based on each user's interest, goals, and behaviors.

2. Optimized user engagement: Based on learnings from analyzing user behavior, they have optimized gamification and incentive models resulting in an improvement in user engagement levels by >15% and a 30% reduction in user bounce rates

3. Correlated insights: Through Munir's data science team, we have been able to unlock new insights into users' lives, helping them understand and predict how their health, lifestyle, sleep, stress and financials are all interconnected.

Razorthink

Razorthink is pioneering Enterprise AI with Deep Learning systems that solve data analysis challenges previously impossible using traditional analytics or data science technologies. Razorthink developed a ‘Last Mile’ hybrid AI system for a leading insurance company in India to drive superior decisions and ideal outcomes for the organization.

The Challenge: The insurance provider had the lowest persistency ratio for fifth year policy renewals in comparison to other top insurance companies. In addition, their traditional machine learning models were becoming too difficult to maintain.

The Solution: Razorthink developed a hybrid Deep Learning model that identified behavior-based micro-segments of users automatically and predicted the propensity of a customer to renew their life insurance policy. The Razorthink solution provided insurances representative with propensity scores for each individual customer as well as an explanation of how the algorithm came up with the score and supporting points.

The Result: The Deep Learning model beat the GINI score by 15% in comparison to the insurance provider’s existing SAS model. Using this unique combination of information, insurance representatives were able to focus efforts on the customers most likely to not renew their insurance coverage during the fifth year. Razorthink’s superior predictive intelligence contributed to $15 Million in incremental revenue for the insurance company over only a few months’ time.

Cogito AI and MetLife

Under Rahul Magan’s leadership, MetLife has deployed Cogito's AI solution. Cogito performs live analysis of voices in phone conversations and delivers guidance, helping agents build deeper emotional connections with customers. It also provides an instant measure of customer satisfaction.

Kanetix and Integrate.AI

To improve customer experience and maximize customer acquisition, Kanetix Ltd implemented an accelerated Conversion Rate Optimization (CRO) program in 2017. Weekly hypotheses, ranging from visual hierarchy tests to psychological principal theories, were evaluated through a variety of A/B and multivariant designs to determine statistical significance.

To further enhance this CRO program, Kanetix.ca then partnered with Integrate.ai to use Artificial Intelligence and Machine Learning to determine the likelihood of a customer to transact so that their experience could be personalized in real time to increase that likelihood. The goal was not to focus on users who are unlikely to convert or those who are most likely to convert, but those in the middle and ‘on the fence’.

Using a large set of Kanetix anonymized data, a model was created providing a predictive score on the likelihood of customers to convert, allowing the targeting of ‘fence-sitters’. Customers were then presented different treatments of the conversion page to nudge them to convert. The data and customer behaviours were then fed back into the Machine Learning algorithm to help improve the overall accuracy of the model.

Over the course of the experiment, Kanetix.ca experienced a conversion increase in auto insurance transactions of 13 percent, a greater than 5x improvement from early results, generated 2.3x ROI, and the efficacy of the platform's predictions improved by approximately 38 percent.

Kanetix.ca is now running 100 percent of Ontario auto insurance quotes through Integrate.ai's platform, improving Kanetix’s overall market position as the leader in online aggregators in Canada.

Auto Lenz by Allai

Allai have created an AI capability that redefines the automobile insurance quote experience by completely removing the need for a customer to enter any information to obtain a quick automobile quote achieved just by taking a picture of an automobile, any automobile, from any angle – the Auto Lenz.

In addition, the customer is provided with data about the vehicle to help them with their decision during this life event instantaneously at the snap of a photo, information such as recall, safety rankings, fuel consumption and many other data points.

The AI can be interacted with by the customer via their mobile camera, SMS, Social Networks or directly from any web page on the internet by simply sharing the photo on from the webpage with the insurance ai. This provides direct connection from the customer to the insurance service provider from anywhere in their digital world.

Question Title

* 3. Claims Application: this award recognizes the most exceptional business case on claims over the past 12 months

Sri Ramaswamy/Infinilytics, Inc.

Under Sri Ramaswamy’s leadership, Infinilytics team developed an innovative way to predict if an insurance claim was going to be litigated by combining artificial intelligence (A.I.) system with a human intelligence (H.I.) expert system, saving the customer millions of dollars in their catastrophic claims.

To improve the customer experience, and reduce claims costs, the creation of a system leveraging A.I. and H.I. allowed insurance companies to predict (greater than 85% accuracy rate) and mitigate the severity of claims from hurricanes Harvey and Irma. The earlier the prediction, along with the context of the prediction, better decisions will be made in processing these claims.

The Infinilytics’ smartC™ platform analyzes all historical claims with applied machine learning and an augmented built-in expert system, enabled claims managers to predict and mitigate earlier which claims might end up in litigation.

Onsite™ Solutions, by EagleView

Within the past year, EagleView has paved the way for virtual claims inspection with the introduction of EagleView Onsite™ Solutions. This tool delivers imagery from multiple camera sources in addition to property data and reports directly to the claims adjusters’ desk, eliminating the high demand for extensive field visits after a catastrophic weather event. Using their vast image library of over 350 million images, EagleView can provide pre- and post-imagery to the adjuster, allowing for a deep understanding of the damage seen directly from the storm. This also aids in identifying fraudulent claims.

Working together with Allstate, EagleView developed this new technology in the spring of 2017 to provide a quicker, more efficient closing of claims post-CAT event. This new solution was created on a strict timeline but was completed and utilized during the 2017 hurricane season. Over 10,500 claims went through EagleView OnSite Solutions following the 2017 hurricane season, closing claims and providing homeowners with aid faster than ever before (by 50-60%), while simultaneously keeping customer satisfaction scores up.

Intelligent Water Monitoring Solution, by Eddy Home

Eddy’s Intelligent Water Monitoring Solution is enabled with AI and IoT technology to proactively detect and prevent leaks and water damage to property. The technology tracks all water usage, utilizing behavioural learning to understand normal usage parameters, and alert users when consumption is outside of this norm. The system includes technology with remote and automatic shutoff capabilities at the water main, sensors placed in various areas, and a dashboard/App that demonstrates water usage and alerts.

The system provides coverage not just to single dwelling homes, but to the multi-dwelling unit space, where water damage incidents are higher and can be much costlier. Recently, Eddy completed a major installation in a luxury condo building in downtown Toronto. The building had experienced a previous major water issue and their insurer was threatening to discontinue coverage. They installed 1,800 sensors in suite and in communal areas accompanied with Eddy’s property manager dashboard. Within 60 days of installation, the system indicated 17 different water incidents, such as a leaking boiler room valve, two frozen heat exchangers, and an overflowing toilet. Left unchecked and unmonitored, the damages and costs would have been significant. The system continues to alert property managers to issues both inside suites and all over the building.

In addition to protecting the property, Eddy’s solution provided their insurance company with the confidence to continue their insurance coverage.

LossDetect™, by Munich Reinsurance America, Inc.

Recent increases in road fatalities and crashes have resulted in increased auto loss ratios. Munich Reinsurance America, Inc. (Munich Re) can identify the core causes of loss of these crashes by utilizing both predictive and prescribed analytic techniques. The loss drivers are addressed by applying technology and risk management solutions. Their product, LossDetect™ (patent pending), is a loss analytics tool developed to improve commercial auto results by utilizing text mining techniques on data gathered from clients’ loss portfolios. The interpretation of LossDetect’s results enables them to recommend innovative solutions for the insured.

LossDetect™ was developed by Munich Re’s Smart Mobility team. A single internal statistical modeler developed our program and supporting model. He was assisted by two internal data analytics team members who standardized rules, provided model reviews and tested each iteration. The IT department, Reinsurance Division, and Specialty Markets business units were then consulted, to obtain market feedback.

LossDetect analyzes over 4,000 records in minutes, and after a consistency review, the entire process is completed within a few hours. As of Feb 2018, the tool has been utilized for 23 clients. Total mitigatable losses identified were over $1. 8bn.The model reduced the time and effort for drawing conclusions from the loss data from 15-person days to 0.5-person days.

Mitchell WorkCenter™ Assisted Review

Mitchell WorkCenter™ Assisted Review is the first integrated workflow solution to leverage artificial intelligence for the estimate review process. By using visual computing to analyze photos,Mitchell WorkCenter™ Assisted Review uses machine-learning technology to recognize incorrect replace or repair decisions, helping clients to minimize estimate errors and maximize reviewer efficiency.

Mitchell WorkCenter™ Assisted Review leverages visual computing technology and promises to improve workflow and claims outcomes. By utilizing millions of damaged vehicle photos, computers are “trained” to recognize vehicle damage and use computer vision to double-check repair vs replace decisions. This technology helps to quickly identify and surface estimates that are potentially inaccurate, allowing carriers to more easily maintain estimate quality and consistency, be more selective about sending appraisers into the field, and improve cycle times and productivity.

Early case studies using the new solution are demonstrating that approximately 10% of repairable fenders were replaced. Not only does this amount to considerable potential gains for insurers, but this reinforces the belief that artificial intelligence can make the claims process quicker, more cost effective and more efficient.

Synthetic Peer, by Mitchell

Mitchell’s patent-pending “Synthetic Peer” approach uses statistical modeling – rather than industry averages – to benchmark insurance carrier claims performance against claims data that is relevant to their book of business.

The following is a case study of the success of the use of Synthetic Peer.
One Mitchell client, a regional consumer auto insurance carrier, has a much different book of business in terms of vehicle population than the rest of the industry. This meant that benchmarking their performance against the wider industry does not necessarily provide much insight into the success of their claims teams.

By creating a synthetic peer data set that normalizes for geographic distribution, as well as book of business, creating a much more “apples to apples” comparison for the client to benchmark against. By using the synthetic peer as the barometer of success rather than an industry with very different vehicle and geographic characteristics, we were able to provide our client with deep, actionable insights that previously would have been unavailable.

The client was able to see an opportunity to increase their rates of repair for major component panels. Mitchell has been working with their client to offer consultative guidance on improving performance in this regard by leveraging industry best practices. The result is a continuous improvement process with a goal of achieving a more accurate estimate that also results in a safe and proper repair.

AI-powered Information Extraction, by DXC Technology

Typically, Insurance/Re-insurance companies operating globally face a huge challenge in processing enormous amount of emails pertinent to claims. They receive millions of emails in multiple languages per month and they manually process them to extract business specific entities like claim number, contract number, name of an insured person/organization, broker name and date of loss.

DXC Analytics has built an NLP powered entity recognizer that has learnt this business/domain patterns by analyzing 200 GB of historic English emails. This solution is a perfect blend of state-of-the-art sequence modeling technique, proprietary ‘relevant-chunk’ detector coupled with superior feature engineering tailored for the insurance domain.

Their ‘Created Features’ exploit the style of writing, grammatical structure found in the English language and other contextual features. “Self-learning” is another key capability. Whenever the confidence scores of the predicted entities are below a certain threshold, a human intervention is requested to review the email and correct the entities which is then fed back as an input to model. The model thus makes use of this right answer to improve its prediction power for future emails. This solution was evaluated on 40,000 unseen emails and the F1-Score was as high as 0.97 for entities like claim number and was around 0.82 for entities like insured name. Further, this solution can be industrialized into client’s existing business process.

This scalable and platform independent solution drastically reduces the time required to process claim-related emails thereby leading to swift processing of claims and faster response to customers. In addition, cost and time benefits of this solution will help organizations optimize their resources and add value.

Intelligent Claims Assignment Solution, by Claimatic

Claimatic’s Intelligent Claims Assignment Solution is a disruptive innovation that performs the complex task of optimizing thousands of decisions in less than a second, to assign claims to the right resources, ultimately reducing the claims cycle by days. Claimatic is the first and only claims assignment solution on the market to incorporate Robotic Process Automation (RPA), and is designed to enhance, not replace, existing technology.

Claimatic is a bolt-on-brain for claims workflow technology that makes existing systems smarter, more valuable and productive. The platform operates behind the scenes and has no visible impact on an organization’s existing workflow, allowing claims managers to continue doing their jobs in the same way, with no loss of productivity due to learning a new system or workflow.

Claimatic positively impacts every cost component of insurance claims management resulting in reductions in overall cycle time, vehicle fuel expenses, litigation, and fraud expenses. The technology has been found to reduce total claims handling expenses by hundreds of dollars per claim and can assign 250 claims to the proper adjusters in seconds - rather than the 22-27 hours it would typically take to assign the same number. The result is increasingly better assignment decisions that are more efficient and consistent, with increased customer satisfaction.

Field Pros Direct

Field Pros Direct is contributing to the industry by providing the first on-demand adjuster network. Hurricane Matthew was a catalyst for Field Pros Direct. Many thought Matthew could be a worst-case scenario for Florida insurers, but he merely skirted the coast and carriers that prepared for 50,000 claims, received <5,000. Despite the small impact of hurricane Matthew some of the largest adjusting firms were fined six figures or more because they couldn’t engage enough adjusters. What’s worse is that 1000’s of adjusters deployed drove back home with little or no work. Delays finding resources and assigning claims occur daily, but Matthew put a spot light on the problem.

Today there is a solution and it is Field Pros Direct. The platform combines their automated adjuster assignment engine with an aggregated network of 1,000s of licensed and experienced adjusters. It utilizes predictive analytics, AI, machine learning, and real time adjuster availability to match each claim with the best available field pro.

This provides improved efficiency in assigning claims, unrivalled CAT capacity, elevated quality, improved cycle times, increased geographic spread, and reduced drive time expense. Our adjusters can handle more claims they prefer in territories they want to cover

Daisy Intelligence

Daisy Intelligence provides an adaptive fraud detection service, rather than a static fraud detection system! Daisy deals with social networks and not just abuse alerts.

Their tool, Daisy Suspicion Index (DSI), is a comparable tool which measures the likelihood that a transaction/entry is money laundering relative to all other transactions/entities in the system. It considers multiple methods of analytical detection to determine which transactions are more likely to be money laundering and it is automatically adaptive to changes in Data without any human supervision. It increases the likelihood of detection and lowers false positive rate. Daisy A.I. can better identify and prevent fraud by finding more patterns to establish risk. They score every attribute of each claim and ‘score’ them against peer group attributes.

Computers put in the same effort analyzing the first claim as the 10,000,000th claim. The result is significant reduction in false positive rate, doubling the recovery rate per hour spent, prevent organized crime and allows for more online processing without increased fraud risk. Their analytic techniques include ‘Probabilistic’ rules to define specific rules to stop known fraud models.

CCC Information Services

CCC is turning the insurance claims process on its head. By leveraging innovation, such as artificial intelligence (AI), telematics and deep learning, they are transforming the traditional claims path into a faster, easier and more efficient process.

Leveraging telematics data, CCC is the first company to automate the First Notice of Loss, seamlessly pushing crash data from a driver’s vehicle into the insurer’s claim system. Instead of waiting for a phone call from a customer, the insurer is instantly notified of a collision and can proactively contact the driver to determine next steps.

With the launch of CCC Smart Total Loss, a single photo can be used to predict whether a vehicle is repairable or a total loss. Instead of waiting for an estimate, insurers can determine within seconds where to route the vehicle for repair or salvage.

Using advanced photo analytics and AI to automate damage detection, CCC’s proprietary heat maps provide a color-coded overlay on a photo of a damaged vehicle to easily detect damage scope and severity. Vehicles routed for repair are processed faster as estimators can accurately inspect vehicle damage from their desk.

CCC is driving a claims revolution, and the disruption has only just begun.

Claim Genius

Claim Genius’s vision is to make “Touchless Claims A Reality”, by using the powerful proprietary AI image analytics platform Claim Genius is developed to create a mobile claim reporting and instant appraisal system which would enhance an automobile insurance carrier’s claim handling process, reduce claim administrative cost, while creating a superior customer claim experience.

Claim Genius’s AI application allows the Insurer to receive critical claim information at the first notice of loss by:

- Simplifying the process to report a claim

- Guiding the customers through its self-guided mobile App to attach vehicle damage photograph and pertinent vehicle information

- Using the proprietary deep machine learning artificial intelligence (AI) technology it instantly determines the Repairable Vs Total Loss condition for any vehicle

- Providing the insurer with an instant initial repair estimate.

- Providing a predictive salvage value

Overall, CG brings an attractive value propositions for the insurance by:

- Creating an Effective Early Tow Program - that evaluates whether the vehicle is total losses via uploaded photos, expediting the movement of vehicle directly to an auction facility instead of repair shop, reducing multiple tows, unnecessary tear downs, and storage, significantly reducing carrier’s loss adjustment expense (LAE).

- Providing Cutting Edge Mobile & AI Technology – which clients can use to attract and retain increasingly millennial and Gen Y customers who demand fast and efficient digital interactions with carriers. CG’s solution allows clients to quickly differentiate their claims process from competitors.

- Integrating Quickly Into Carrier Tools – their suite of APIs and tools make it easy for carriers to plug CG’s technology into their existing workflows and systems and go to market rapidly.

- Quicker resolution of the claim which leads to higher customer satisfaction and customer retention post loss.

The Claim Genius application can be tailored to the needs of the individual carrier.