Using Generative AI at Work, with Method and Safely

Using Generative AI at Work is a beginner-level course of 5 lessons (about 1 h 25 min), with 15 exercises and 15 quiz questions. No prerequisites. Exercises run in the browser, with nothing to install.

  • Generative AI
  • Beginner
  • 5 lessons
  • 1 h 25 min
  • 15 exercises
  • Exercises run in the browser

The course in brief

This course is for any employee who wants to use a generative AI assistant in a useful and safe way. You will understand what a language model can and cannot do, learn to write precise requests, check the answers, and protect company data. Finally, you will apply a complete method to writing, summarizing, and analysis tasks. No technical skills are required.

What you will be able to do

  • Explain in simple terms how a language model works and what its limits are
  • Write structured prompts and improve an answer through iteration
  • Check facts, figures, and references before using an answer
  • Protect confidential and personal data when using an assistant
  • Apply a prepare, ask, verify, adapt, own cycle to workplace cases

Syllabus: 5 lessons

  1. 01

    What an LLM Can and Cannot Do

    15 minFree preview
    • Explain in simple terms how a language model produces an answer
    • Identify the tasks where generative AI is useful and those where it is risky
    • Recognize its limits: outdated knowledge, no verification, variable answers

    Language model (LLM)Model limitationsSuitable use cases

  2. 02

    Writing Good Prompts

    20 min
    • Build a complete request: role, context, task, format, constraints
    • Improve an answer through successive iterations
    • Use examples to get the expected style

    Writing promptsIterationExamples in the prompt

  3. 03

    Checking, Citing, and Avoiding Hallucinations

    15 min
    • Set up a verification routine suited to the stakes
    • Reduce the risk of hallucination through the way you ask
    • Transparently disclose the use of AI when it is expected

    Verifying answersHallucinationsTransparency of use

  4. 04

    Confidential Data and GDPR

    15 min
    • Distinguish the information that can and cannot be entered into an AI assistant
    • Apply the basic GDPR rules to everyday AI use
    • Anonymize or generalize a request before submitting it

    Data confidentialityGDPR and AIApproved tools

  5. 05

    Business Cases: Writing, Summarizing, Analysis

    20 min
    • Apply a complete method to three common cases: writing, summarizing, analyzing
    • Define the essential human checks for each case
    • Honestly estimate the real gain from using AI

    Assisted writingDocument summarizationAssisted analysis

Final project

My AI Usage Guide for My Role

Choose three recurring tasks from your job (or from a fictional sales assistant role). For each, build a generative AI usage sheet: benefit, sample prompt, authorized and prohibited data, checks to perform, estimated time with and without AI. Finish with a page of personal usage rules.

Deliverables

  • Three usage sheets (task, complete sample prompt, authorized and prohibited data, checks, estimated gain)
  • An example of an answer obtained and its corrected final version for at least one task
  • A page of personal usage rules (5 to 8 rules)

Assessment criteria

  • Structured prompts (context, task, audience, format, constraints)
  • No personal or confidential data in the examples
  • Checks suited to the stakes of each task
  • Honest estimate of the gain, verification time included
  • At least one task identified as unsuitable or limited, with justification

Frequently asked questions

Do I need to know how to code to take this course?

No. The whole course is done in everyday language: you write requests, analyze answers, and apply good-practice rules.

Is the course tied to a specific tool?

No. The methods apply to any generative AI assistant. For the hands-on exercises, preferably use the tool approved by your company.

Can you trust an AI's answers?

They are useful but can contain errors presented with confidence. The course teaches you to check facts, figures, and references before using them.

Can you enter customer data into an AI assistant?

Only if the tool and the use are authorized by the company, and by limiting the data to what is strictly necessary. In most cases, an anonymized request is enough.

This course, told for your job

Say what you want to be able to do: “Using Generative AI at Work” takes its place in your path, with examples from your field. The concepts stay those of the course.

The preview is free, and you can take the courses one at a time without paying. The path only contains lessons from the catalogue.