Sprout.ai Helping Claim Handlers Look to Future
What does Sprout.ai do?
Sprout.ai validates claims, checks for fraud, reduces waste and abuse, and also identifies outliers. It refines and improves customer data, with GDPR-compliant external information.
What is Sprout zero-shot?
Sprout zero-shot is a new foundational model we’ve developed that closes the gaps in claims processes and increases the efficiency of claims automation. It leverages foundational claim data to review and assess claims without ever having seen a claim of that type before. This enables insurers to speed up claim onboarding times and minimises the data required for training.
So far, the Sprout zero-shot model has already delivered a 50% increase in claim handler efficiency, with a 90% reduction in training data required.
It has an in-built quality-checking function to speed the claims submission process, and natural-language processing advancements to extract and classify data, both from structured and unstructured documents, such as invoices.
Who will Sprout zero-shot benefit most?
In April 2024 we released the results of our survey of claims handlers working in different lines across the UK and US. This gives us insights into the biggest pain points they face and the biggest areas of potential for an AI-led approach.
The research shows claims handlers working in life insurance and travel insurance struggle most with document and evidence processing, with 51% of handlers in both lines reporting this as a challenge. AI can make a huge difference to productivity and customer satisfaction in these lines.
How does Sprout zero-shot use NLP?
The model has a validation engine that uses neuro linguistic programming (NLP) and it works with the extraction engine to facilitate automated checking on all extracted data to ensure accuracy.
This has been trained using multiple languages, so it can process data in whatever form it comes. This is particularly useful in travel claims, where evidence might be submitted in different languages and forms.
Are there any associated risks?
Overall, a zero-shot model for insurance claims can offer efficiency and scalability where there isn’t existing data to train the model. However, due to the use of synthetic data to train the model, insurers need to consider bias and accuracy. Sprout.ai has been able to achieve 96% accuracy rate on claims automation.
How can insurers prepare staff for AI?
Contrary to popular belief, claims handlers aren’t worried about being replaced by AI. Our research shows 95% of UK and US insurance claims handlers are confident technology will significantly impact claims processing in the next five years.
To make sure employees are receptive to AI integration, insurers should remind them AI exists to make their job easier and less stressful, not to take their role away. It’s also worth reminding them of the huge impact AI can have on customer experience, ultimately reducing complaints and improving customer satisfaction.
**************
Make sure you check out the latest industry news and insights at InsurTech Digital and also sign up to our global conference series - FinTech LIVE 2024
**************
InsurTech Digital is a BizClik brand
**************
