Salesforce study warns against rushing LLMs into CRM workflows without guardrails

Salesforce’s CRMArena-Pro test reveals state-of-the-art models—including GPT-4o—fail to meet critical enterprise needs in CRM, especially around secure, multi-turn processes.

A new benchmark study from Salesforce AI Research has revealed significant gaps in how large language models (LLMs) handle real-world customer relationship management (CRM) tasks.

https://www.cio.com/article/4008228/salesforce-study-warns-against-rushing-llms-into-crm-workflows-without-guardrails.html