While the world is going ga-ga over Large Language Models like ChatGPT and Bard and what they can do, the accounts receivable management industry has largely been on the sidelines and slow to adopt the technology, over fears that information put into the models can then be shared with others around the world, creating significant third-party disclosure risks. A company yesterday announced the release of a Large Language Model for the contact center industry, which was trained on redacted data and stores customer account data in dedicated storage partitions.
Observe.ai, the company behind the model, said that its LLM is built to improve call center agent performance across multiple communication channels and types of interactions and was trained using real-world contact center interactions between consumers and agents. Companies using the LLM from Observe.ai will be allowed to tailor their version of it to suit their specific needs, the company said.
“We have a nuanced and accurate understanding of what ‘successful’ customer experiences look like in real-world contexts,” said Swapnil Jain, Observe.ai’s chief executive officer, in a published report. “Our customers can then further refine and tailor this to the unique needs of their business. Our approach provides a full framework for contact centers to calibrate the machine and verify that the actual outputs align with their expectations. This is the nature of a ‘glass box’ AI model that offers complete transparency and engenders trust in the system.”
In testing against ChatGPT-3.5, Observe.ai’s LLM was 35% better at summarizing conversations between contact center agents and consumers and 33% better at analyzing consumer sentiment, according to the report.
The model includes coaching features for contact center agents, as well as a tool that facilitates responses to customer inquiries using a company’s internal knowledge bases, and a tool that summarizes conversations and automates the post-call note process.