Dimitri Masin, CEO of Gradient Labs, critiques the current customer support pricing model where businesses pay per interaction regardless of success. He highlights that this approach does not incentivize companies, like Salesforce, to improve the performance of their AI agents. With statistics showing low success rates for multistep queries, Masin advocates for a billing system where companies only pay for resolved issues, thus tying costs directly to service quality. This method addresses ROI concerns, especially amid rising compute costs in AI development.
Masin argues that companies using AI agents for customer support should only pay when the bot does its job, creating operational efficiency and incentive for improvement.
The problem with the current pricing model is that it doesn't create any incentive for Salesforce to actually enhance its AI capabilities.
Collection
[
|
...
]