Understand generative AI fundamentals

Official references for this objective

• Microsoft — Generative AI | Microsoft Learn

  They can process and generate natural language, but they require more components to handle user interactions, security, and other functionality to respond to or act

• Microsoft — Monitoring Generative AI applications | Microsoft Learn

  Ensuring model accuracy and reliability: Models evolve, and with evolution can come drifts in accuracy.

• Microsoft — Security planning for LLM-based applications | Microsoft Learn

  Reduce the number of extensions to minimum necessary. Limit the functions of an extension to only required functionality.

• Microsoft — Evaluation metrics | Microsoft Learn

  Faithfulness: Measures the factual consistency of the generated answer against the given context. If any claims are made in the answer that cannot be deduced

• Microsoft — Safeguarding LLM security & safety evaluations | Microsoft Learn

  Prompt Shields – Preventing adversarial prompt inputs (prompt injection) from manipulating LLM behavior.