The Government is developing a ChatGPT-like chatbot hotline that can be used to handle consumer complaints. The gadget will also be built so that users may communicate with it through audio messages in a variety of Indian languages.
To be clear, the chatbot that is now being developed is distinct from the one that was conversationally deployed last month to address customer concerns. This is due to the fact that it requires the user to choose from a list of possibilities in a step-by-step fashion, starting with state, city, industry, brand, issue, product value, etc., as opposed to allowing him to just enter facts as in a typical discussion.
According to sources, the Government’s Bhashini platform will handle the language translation aspect of the new tool being developed, but the challenge is finding a large language model (LLM) that can quickly comprehend problems connected to customer complaints.
A conversational AI chatbot’s LLMs process your prompts when you communicate with it. They aid the AI in comprehending your material, dissecting its meanings, and establishing context in order to provide a response.
“We can’t use the large language models (LLM) that many firms have already developed. An LLM with knowledge of the particular subtleties of consumer concerns and legislation would be required for this “a person with knowledge remarked.
Since they have more exposure to the language used in that particular domain, LLMs educated on domain-specific data are better able to comprehend the context and subtleties of that particular area.
As opposed to a language model trained on general language data, it is understood that a language model trained on medical data would be better at comprehending medical jargon and terminology and would be more accurate in responding to medical-related inquiries. In a similar vein, compared to a language model trained on general language data, a language model trained on legal data would be better at comprehending legal terminology and ideas and would be more accurate in answering legal-related queries.
The amount and quality of the training data, the model’s architecture, and the effectiveness of the pre-and post-processing techniques employed are all critical aspects that affect how well a language model performs.
Furthermore, a significant issue with conversational AI chatbots like OpenAI’s ChatGPT and/or Google’s Bard has been what is known as “hallucinations” – these chatbots have a tendency to fabricate stories when confronted with inquiries about details or context they are unfamiliar with.
