Salesforce study finds LLM agents flunk CRM and confidentiality tests

go.theregister.com/feed/www.theregister.com/2025/06/16/salesforce_llm_agents_benchmark

6-in-10 success rate for single-step tasks
A new benchmark developed by academics shows that LLM-based AI agents perform below par on standard CRM tests and fail to understand the need for customer confidentiality.…

This story appeared on go.theregister.com, 2025-06-16 13:19:11.
The Entire Business World on a Single Page. Free to Use →