Stanford Secure GPT: Smart Prompting in a Secure AI Sandbox

Whether you're an AI newbie or someone curious about integrating secure AI into your daily work, this hands-on workshop will equip you with the knowledge and skills to leverage SHC-SoM Secure GPT confidently and responsibly.

We'll break down the essentials of prompting, explore reasoning strategies, and demystify what's under the hood of this powerful tool.

Through practical exercises, you'll learn how to structure your requests to generate meaningful and safe outputs -- whether you're summarizing documents, brainstorming ideas, or simplifying complex concepts. 

Learning Objectives
By the end of this workshop, participants will have the opportunity to: 
- Understand the role and limitations of Secure GPT within Stanford Medicine's ecosystem.
- Identify key features of effective prompt design for productive AI output.
- Apply secure prompting techniques to solve common workplace challenges.
- Analyze differences among AI platforms (OpenAI, Google, Meta, Anthropic, etc.) for various use cases.
- Evaluate when and how to responsibly use Secure GPT with sensitive information.
- Design custom prompts tailored to different task types and personas.
 

Topics to Be Covered
- Introduction to Secure GPT (SHC-SoM): What it is, how it works, and why it matters
- Prompting Fundamentals: Clarity, context, structure, and personas
- Reasoning with AI: How to guide complex queries and ensure relevance
- Platform Deep-Dive: Quick comparisons between Secure GPT and tools from OpenAI, Google, Meta, DeepSeek, and Anthropic
- Security Best Practices: PHI/PII handling, what not to share, and Secure GPT guardrails
- Creative vs. Precise Modes: When to use each and how to optimize results
- Hands-On Practice: Real-world scenarios for document summarization, brainstorming, and data-friendly prompts
- Audio and Speech-to-Text Use Cases: Where Secure GPT shines with uploaded formats
- Common Pitfalls and FAQs: Misconceptions, token talk, and troubleshooting tips

Special Features 
Practical, Hands-On Experience:
Task-based learning: Participants will have an opportunity to practice tasks that simulate real-life scenarios, encouraging the application of AI tools to solve various everyday challenges.