# FAST-03 Reading List: Prompt Engineering for the Workplace Provider: 7formations | Course: FAST-03 | Instructor: Wadood Chaudhary | Links verified by fetch or search as of Oct 2026 Readings come mainly from the vendors' own prompting guides, which are free, current, and written for the tools trainees use, plus a small set of risk and data-protection sources. Vendor guides overlap; read the one for your employer's tool first. **How to use.** Required readings take 20 to 45 minutes per session and are done before (or right after) the session they support. Optional readings go deeper. 'Living page' means the publisher updates it continuously; read the current version. Reading times are estimates for the parts that matter to this course, not the whole document. --- ## Session 1: A prompt is an assignment ### Required - **Get started writing prompts in Microsoft Copilot** Publisher: Microsoft Support | Date: Living page | Reading time: about 10 min URL: Why read: Microsoft's prompting guidance (goal, context, source, expectations), which matches the five-part card. Read this one first if you use Copilot. - **Prompt engineering best practices for ChatGPT** Publisher: OpenAI Help Center | Date: Living page (updated October 2026) | Reading time: about 12 min URL: Why read: OpenAI's prompt best practices: put instructions first, separate them from context with delimiters, be specific about format. The vendor version of 'fence the source'. - **Reduce hallucinations** Publisher: Anthropic (Claude Docs) | Date: Living page | Reading time: about 10 min URL: Why read: Practical techniques to reduce invented content: allow 'I don't know,' ground answers in quotes. This is the reasoning behind 'not in source.' ### Optional - **Google Workspace with Gemini Prompt Guide** Publisher: Google Workspace | Date: Living page (guide updated periodically) | Reading time: about 25 min URL: Why read: Google's Gemini for Workspace prompting guide, with role-by-role examples. Skim the section for your role. - **Prompt engineering overview** Publisher: Anthropic (Claude Docs) | Date: Living page | Reading time: about 20 min URL: Why read: Anthropic's prompt engineering overview. The pages on being clear and direct and using XML tags explain fencing and structure in depth. --- ## Session 2: Passes and critique ### Required - **Effective Prompts for AI: The Essentials** Publisher: MIT Sloan Teaching & Learning Technologies | Date: Living page | Reading time: about 10 min URL: Why read: MIT Sloan's short guide to effective prompts, including iterating and refining rather than expecting one perfect prompt. ### Optional - **Plain language guide** Publisher: Digital.gov (U.S. General Services Administration) | Date: Living page | Reading time: about 20 min URL: Why read: Plain-language guidelines for pass 3: write for the reader, put the main point first. - **Why language models hallucinate** Publisher: OpenAI | Date: September 5, 2025 | Reading time: about 12 min URL: Why read: OpenAI's explanation of why models guess rather than abstain. Useful context for why later passes can add 'helpful' facts. - **When AI Gets It Wrong: Addressing AI Hallucinations and Bias** Publisher: MIT Sloan Teaching & Learning Technologies | Date: Living page | Reading time: about 10 min URL: Why read: Addressing AI hallucinations and bias; reinforces checking each pass. --- ## Session 3: Quality gates and your library ### Required - **201 CMR 17.00: Standards for the Protection of Personal Information of MA Residents** Publisher: Mass.gov, Office of Consumer Affairs and Business Regulation | Date: Regulation page dated 10/19/2017 (in force since 2010) | Reading time: about 15 min URL: Why read: 201 CMR 17.00, the Massachusetts data-security regulation. Read the definition of personal information (17.02) to know what always goes on the strip list. ### Optional - **OWASP Top 10 for LLM Applications** Publisher: OWASP Gen AI Security Project | Date: 2025 edition | Reading time: about 20 min URL: Why read: The OWASP Top 10 for LLM applications. Read 'Sensitive Information Disclosure' and 'Prompt Injection' for why pasted content and shared prompts need care. - **Data, Privacy, and Security for Microsoft Copilot** Publisher: Microsoft Learn | Date: Living page | Reading time: about 10 min URL: Why read: How Microsoft Copilot handles prompts and data in work accounts. Check the equivalent page for your tool to understand what happens to what you paste. - **Enterprise privacy at OpenAI** Publisher: OpenAI | Date: Living page | Reading time: about 8 min URL: Why read: OpenAI's business data privacy page (ChatGPT Business and Enterprise). Compare it with the Copilot page. - **Attorney General Advisory on the Application of the Commonwealth's Consumer Protection, Civil Rights, and Data Privacy Laws to Artificial Intelligence** Publisher: Massachusetts Attorney General's Office | Date: April 16, 2024 | Reading time: about 15 min URL: Why read: The Massachusetts Attorney General's April 2024 advisory: existing consumer-protection and data-security laws apply to AI. Context for 'the sender owns it.' --- ## Link verification Links verified by fetch or search as of Oct 2026. Links that could not be confirmed were dropped and are not listed. The instructor re-checks links before each cohort. **Dropped during verification (not listed above):** - Microsoft Learn module "Craft effective prompts for Microsoft 365 Copilot" (404)