How Salesforce uses AI

Salesforce's Sales Cloud engineering team built BugWiser, which pairs a custom machine learning model trained on years of historical bug classifications with a large language model (Anthropic's Opus). The model classifies each customer reported bug with a confidence score and the evidence behind it, and the LLM writes the engineering context and summaries. High confidence predictions can be accepted automatically, lower confidence ones go to manual review, and engineer corrections retrain the model. Salesforce reports the first Sales Cloud and Service Cloud deployments.

Salesforce, with about 80,000 employees, governs about 102,000 pay decisions a year, and Syndio's platform manages the complexity of that analysis across more than 100 countries. Benioff first took a public stand on pay equity in 2015. Salesforce initially ran pay equity analysis in house with a model built by its own data science team, and later replaced it with Syndio's platform as the company scaled.

Salesforce's internal IT team, Techforce, handles more than 25,000 requests a month. It built the Techforce Agent on Agentforce and Service Cloud and put it in Slack, where employees describe an issue and get an answer, often a link to the self service password reset portal, or an escalation to a human with the conversation attached. The CIO reports 28,000 employees using it. In this version the agent surfaces knowledge and guides to self service; direct action on the employee's behalf is named as the next version. Salesforce's customer story adds that the agent had autonomously resolved more than 9,500 support requests in Slack, saving nearly $57,000 in support costs.