To main content

Enabling Flexible Conversational AI by Integrating Large Language Models (FlexCon)

The project explores how large language models (LLMs) can be integrated into conversational AI for customer service in a safe, responsible, and user-friendly manner.

Contact person

Customer service in the banking, insurance, and telecom sectors is increasingly provided through traditional conversational artificial intelligence (AI). While conversational AI is helpful for many customers, the technology can also lead to frustration if perceived as too rigid or generic. This presents a significant challenge for companies, whose mission is to deliver high-quality customer service through conversational AI.

Enter the world of LLM like GPT! These advanced models are known for their incredible ability to provide flexible and personal responses on various topics. In this IPN project, we are looking to harness the power of LLMs to revolutionize conversational AI for customer service. The goal? By integrating LLMs into the Boost ai platform, we aim to make conversational AI smarter, more flexible, trustworthy, and user-friendly. 

This is easier said than done, and the challenges are many. For example:

Which use cases are LLM best suited for? How can we ensure that they are reliable? And not least, how can we monitor and understand the impact this technology has on customers?

SINTEF is the project coordinator and contributes valuable user insights through surveys, interviews, and prototype testing.

 

Key facts

Project duration

2024 - 2027

Funding

The Research Council of Norway

Partners

Boost ai, Telenor, Sparebank 1 Sør-Øst Norge, and SINTEF

Explore research areas