Which feature in the Einstein Trust Layer helps to minimize the risks of jailbreaking and prompt injection attacks?

The correct answer is C. Prompt Defense. The Einstein Trust Layer includes three key security features. Secure Data Retrieval and Grounding ensures AI responses are anchored to your org's data. Data Masking anonymizes sensitive PII before it leaves Salesforce. Prompt Defense is the specific feature designed to protect a

AI Features for Service (e.g., Einstein Bots, Next Best Action)

Question

Options

ASecure Data Retrieval and Grounding
BData Masking
CPrompt Defense

How the community answered

(47 responses)

A
6% (3)
B
2% (1)
C
91% (43)

Explanation

The Einstein Trust Layer includes three key security features. Secure Data Retrieval and Grounding ensures AI responses are anchored to your org's data. Data Masking anonymizes sensitive PII before it leaves Salesforce. Prompt Defense is the specific feature designed to protect against adversarial inputs - it detects and blocks jailbreaking attempts (trying to override AI instructions) and prompt injection attacks (malicious instructions hidden in user content or retrieved data) before they can manipulate the AI model's behavior.

Topics

#Einstein Trust Layer#Prompt Defense#AI Security#LLM Security

Community Discussion

No community discussion yet for this question.

Full AI-201 Practice