Optimize customer support workflows with generative AI¶

Access this AI accelerator on GitHub

Customer support is a crucial part of every organization. With the proliferation of social media, customer-centric organizations are using social media platforms to provide customer support. Irrespective of the platform, support automation has been actively pursued by organizations to improve the customer experience and loyalty while reducing the workload on support teams.

While automated prioritization and routing have been solved using predictive models, automated resolution is still an active area of research. The majority of support requests are primarily requests for information (level one requests). Handling these requests and would benefit from automation.

In this accelerator, generative AI models cater to level-one requests, allowing support teams to focus on more pressing and high-visibility requests. Learning from historical communications, generative AI responses can maintain the same standard of support communication that customers are used to.

Because use of generative AI comes at a cost—per request, compute time, per token—costs can quickly balloon. To mitigate this problem, you can use predictive ML models for routing standard requests to the generative AI model and high-priority severity requests to human staff.