Review Generative AI Foundational Platform Security

Learn about Generative AI foundational platform security.

• LLM Data Masking (Available for Non-Agentforce Generative AI) Control
  Automatically identifies and masks sensitive data (PII, PCI, and so on) within a prompt before it is transmitted to an external Large Language Model (LLM).
• Safety and Security - Toxicity Detection Control
  Automatically scans AI prompts and generated responses to identify, flag, and score harmful language across multiple categories (for example, hate speech, violence, profanity).
• Safety and Security - Prompt Injection Detection Control
  Detects and mitigates prompt injection attacks where users attempt to overwrite system instructions to force the AI into unintended or malicious behaviors.
• Einstein Audit, Analytics, and Monitoring Setup Control
  Centralizes the logging of Generative AI prompts, LLM responses, and user feedback into Data Cloud for compliance, auditing, and model optimization.
• Prompt Builder Usage, and Feedback Metrics Control
  Monitors the adoption and quality of specific Einstein prompt templates by tracking generation counts and user-submitted feedback.
• Einstein - Agent Analytics Control
  Activates the continuous collection and visualization of autonomous agent interactions, enabling performance tracking and ROI measurement.
• Einstein Audit and Monitoring - Agentforce Session Tracing Control
  Records the step-by-step logic, thought process, and tool-calling sequence of autonomous agents for every interaction.