Student Goals Agent Considerations

Non-Deterministic Nature of AI

The Student Goals Agent uses generative AI, so it may produce varied responses for the same input, even within the same session. This variability is an inherent feature of large language models. It is crucial to communicate this to students upfront so they understand that the agent's suggestions aren't fixed or absolute. This sets realistic expectations and prevents reliance on a single, definitive answer.

Responsible Use of Student Data

The agent leverages specific student attributes to provide personalized recommendations, including but not limited to:

• Academic Data: Major, GPA, academic standing, year of study, first-generation status, and dual-degree enrollment.
• Engagement Data: Membership in campus groups and involvement activities.
• Self-Reported Information: Results from career, preference, and personality assessments.

Before deployment, universities must:

• Ensure Transparency: Clearly inform students what data is being used, why it's being used, and how it informs the agent's recommendations.
• Update Disclosures: If any modifications are made to the prompt grounding flow, the data use policy must be updated to reflect these changes.
• Provide Control: Give students the option to opt out or seek clarification on their data's use.

Recommendations Are Guidance, Not Directives

Remind students that the agent's suggestions are intended to be a starting point for exploration, not a prescriptive plan.

• They aren't obligated to accept any recommendation
• Critical thinking is encouraged, and they should evaluate the relevance and value of all suggestions.
• The agent isn't a substitute for human advisors, mentors, or counselors. Encourage students to always discuss significant goals and plans with human experts.

Establish Clear Usage Boundaries

To ensure the agent is used effectively and safely, universities should establish and communicate clear guidelines.

• A concise onboarding guide that explains the agent's purpose.
• Examples of appropriate queries and use cases.
• A clear statement that the tool is a supplemental aid, not a replacement for human guidance.

Handling Sensitive and Vulnerable Interactions

Given its direct interaction with students, the agent may encounter sensitive topics, such as:

• Self-harm or suicidal ideation
• Inappropriate, aggressive, or concerning language
• Mentions of emotional distress

Universities should proactively manage these risks.

• Enable audit trails: Monitor usage patterns, tag frequencies, and trace individual sessions in the Command Center.
• Implement Custom Safety Workflows: While Agentforce has a built-in TRUST layer, institutions should create their own custom workflows to address specific campus needs.
• Define Clear Escalation Protocols: Establish a clear protocol for routing high-risk interactions to appropriate human support, such as a campus mental health team or crisis counselor.
• Communicate AI's Limitations: Be transparent with students that the agent cannot provide professional help in crisis situations.

Promoting Student Awareness and Informed Use

To empower students to use the agent responsibly, encourage them to:

• View the agent as a co-pilot, not a final authority
• Seek diverse human perspectives when interpreting and acting on AI recommendations
• Use the tool as a starting point for deeper advising conversations with academic advisors, career counselors, and mentors

Language and Locale Support

Student Goals Agent supports the languages and locales supported on the Salesforce generative AI platform, as described in Einstein Generative AI Supported Languages and Locales.

Large Language Model Support

Student Goals Agent supports the models supported on the Salesforce generative AI platform, as described in Large Language Model Support.

Einstein Trust Layer Service Support

Student Goals Agent supports the Trust Layer services provided on the Salesforce generative AI platform, as described in Einstein Trust Layer. Ask your system administrator about which Einstein Trust Layer services are enabled in your org and available for Student Goals Agent.

Usage Types Billed By Student Goals Agent

Student Goals Agent uses generative AI and Data Cloud to ingest, store, and process data. You’re billed in these categories when you use Student Goals Agent. Before deployment, work with your Salesforce account team to confirm license availability and plan credit usage.

Tip This feature has access to Digital Wallet, a free account management tool that offers near real-time consumption data for enabled products across your active contracts. Access Digital Wallet and start tracking your org’s usage. To learn more, see About Digital Wallet.

When you use Student Goals Agent, credits can be consumed in usage types on other consumption cards by related features such as Agentforce Data Library or Audit and Feedback.