Mobile AI Claims solution for Insurers
Key Challenges
Key Results
Overview
Mobile & AI Claims Solution for Insurers
MotionsCloud helps property and vehicle insurers to streamline and automate claims processes through mobile self-service, artificial intelligence (AI computer vision technologies) damage evaluation, and live video inspection technologies modules that can plug into existing claims systems. The solution reduces claims cycle time from days to hours, reduces claims processing cost up to 75%, and improves the customer claims experience. Currently, the MotionsCloud is serving clients in the US, Europe, and Southeast Asia.
Challenges
Currently, the MotionsCloud client is using vision technologies for live inspections to fasten up the claim process. They wanted to develop a solution that should detect false insurance claims when a photo-shopped or altered image is uploaded. The client wants to build a scalable infrastructure to analyze 10,000+ images a month stemming from around 1000 insurance claims to determine if any of the images have been edited.
Solution
- Ankercloud has developed & trained a model on AI with AWS cloud infrastructure. The model was built using services like Sagemaker and Lambda to identify the fraudulent images from the pool of images.
- In order to make the process easier and more convenient, and UI is built where the images can be uploaded and services like Amazon S3 are used to store the result as well as the image.
- Following that, Ankercloud helped with containerizing the client’s ML training and testing codes using Amazon Elastic Container Service. This helped to stabilize application performance at scale and improve operational efficiency.
Business Outcome
- The speed and accuracy of detection of the fake images were drastic and the client was able to cut down the processing time to 80% and achieved high accuracy.
- The solution also provided automation to support the growing businesses without increasing resources.
