# Generative AI: understanding and using AI in your job

URL : https://skillrung.com/en/courses/artificial-intelligence/generative-ai-at-work
Topic: Artificial intelligence · Duration: 60-80 min · Updated on 2026-09-11

## In brief

Generative AI produces text, images and tables from a written prompt. It does not look anything up in a source of truth: it writes the most likely continuation of what came before, which makes it fast and sometimes wrong. At work, three habits are enough: give it a complete context, never upload data that identifies a person, and reread every output before you sign it with your own name.

## Programme

- 1. Why generative AI is coming into your work : Identify what generative AI changes in your work and what it does not change
- 2. What a model really does : Explain in three sentences what a language model produces and by what mechanism
- 3. Write a prompt that works : Build a complete prompt and follow up usefully when the result is not right
- 4. Your data, and what leaves your premises : Decide what you entrust to an assistant, to which service and under which contract
- 5. Check before you sign : Trigger the checking that a model never does on its own, and document the control
- 6. Fit AI into your week : Spot the tasks where AI really helps and those where it has no place
- 7. Your action plan : Choose three uses and fit them into your week without disrupting anything

5 content sections, each with its quiz. Three level quizzes: Beginner, Intermediate, Expert.

## Free preview (first 4 slides)

### 1. An assistant that speeds everything up, mistakes included

Before a New York court, two lawyers filed a brief citing six court decisions that did not exist. A conversational assistant had produced them, complete with the judges' names and the page numbers. Nobody had checked. The court sanctioned the lawyers. The tool had not lied: it did what it is built to do, produce plausible text. Tomorrow, the same mechanism will produce an equally convincing memo, straight out of your logiciel métier.
  - A federal court in New York sanctioned two lawyers who had filed a brief citing six non-existent court decisions, produced by a conversational assistant. (Justia : US District Court, Southern District of New York, 2023-06-22, https://law.justia.com/cases/federal/district-courts/new-york/nysdce/1:2022cv01461/575368/54/)

### 2. What you will be able to do by the end

- Write a prompt that gives a usable result the first time
- Decide whether données personnelles can go into the assistant
- Check an output before you sign it with your name

### 3. Three questions before you upload anything

Ask yourself these before uploading a document to an assistant, at work or at home, every single time.
- Is it confidential? Think of données personnelles, and of prices
- Is it contractual? A client may forbid you from using third-party tools
- Is it checkable? Would you spot an error in the answer

### 4. Where do you stand?

Three questions to gauge how you actually use it, based on what you write every week to collègues et partenaires, not on textbook cases.
_Self-assessment of 3 questions in the player._

## Access

The first four slides of every course and the Beginner-level quiz are free, with no account. From the fifth slide onwards a subscription is required: €29.99/month incl. VAT, no commitment, cancellable online. skillrung is not a French training body certified under the Qualiopi scheme; its content is not eligible for the French CPF, OPCO or DPC schemes.

## Frequently asked questions

### Do you need to know how to code to use generative AI at work?

No. Everything is done in plain language, by writing what you want to get. The skill that counts is not technical: it consists of describing a need precisely, then rereading what comes out. This course requires no IT knowledge and does not ask you to install any software.

### My employer has not given me any tool. What can I do?

Start by asking which tool is authorised and under which contract. Until the answer is in writing, assume that no data from your work should be uploaded to a consumer service. You can, however, train yourself, prepare your prompts and identify the tasks that would save time.

### Can what I type be used to train a model?

On a consumer account, it depends on a privacy setting that only you can change, and the default value varies from one service to another. On a business plan, the contract forbids it and the administrator guarantees it for the whole organisation. When in doubt, assume the answer is yes.

### How can I tell an answer is wrong when I do not know the subject?

You will not, and that is precisely the risk. The safeguard is never to use an output on a subject that nobody around you masters, and to systematically ask for the source and the date of every figure. An answer without a verifiable source does not get signed.

### Will AI eliminate my job?

This course makes no prediction. What can be observed today is a shift: repetitive production tasks leave, time spent reviewing and deciding increases. That shift calls for more experience, not less, because you have to spot a result that is credible but false.

## Sources

- Règlement (UE) 2024/1689 établissant des règles harmonisées concernant l'intelligence artificielle : EUR-Lex : Journal officiel de l'Union européenne, 2024-07-12, https://eur-lex.europa.eu/eli/reg/2024/1689/oj?locale=fr
- Recommandations de sécurité pour un système d'IA générative : ANSSI, 2024-04-29, https://messervices.cyber.gouv.fr/guides/recommandations-de-securite-pour-un-systeme-dia-generative
- Les fiches pratiques IA : CNIL, 2026-09-11, https://www.cnil.fr/fr/les-fiches-pratiques-ia
- Mata v. Avianca, Inc., opinion and order on sanctions, No. 22-cv-1461 (S.D.N.Y.) : Justia : US District Court, Southern District of New York, 2023-06-22, https://law.justia.com/cases/federal/district-courts/new-york/nysdce/1:2022cv01461/575368/54/