Training details
Generative artificial intelligences are profoundly transforming the ways we work, analyze, and produce content.
This training course enables you to acquire the key prompting skills needed to interact effectively with these tools, understand their limitations, and use them in a reliable, responsible, and controlled manner.
Training objectives
By the end of this training, participants will be able to:
Understand how generative AI works and what its limitations are
Identify relevant use cases, biases, and risks in order to take a critical view of the answers produced.Write clear, structured, and effective prompts
Design precise instructions that include context, constraints, tone, and format to obtain usable results.Use advanced prompting techniques for complex cases
Apply role prompting, iterative prompting, example-based prompting (few-shot), and structured prompting to improve response quality.Adopt a professional, reliable, and responsible use of AI
Assess the reliability of outputs, prevent hallucinations, and address security, ethical, and compliance requirements.
Training content
Introduction to generative AI
What is a language model and how does it learn?
Model limitations and biases (hallucinations, training data, context).
Types of use cases: writing, summarization, ideation, translation, coding, etc.
Prompting fundamentals
What is a prompt?
Structure of a good prompt:
Context (who, what, for whom, why)
Clear, action-oriented instruction
Constraints or desired output format
Tone, language level, style
Common mistakes to avoid (vagueness, overload, lack of framing).
Advanced prompting techniques
Role prompting (“You are an expert in…”)
Progressive prompting (iterative, conversational).
Simplified chain-of-thought: reasoning step by step without overloading the prompt.
Few-shot prompting: providing examples of the expected answer.
Structuring prompts for complex tasks (tables, summaries, reports).
Response checking and reliability
How to spot “hallucinations” or incorrect answers.
Validation techniques:
Check sources (citations, references, internal consistency).
Rephrase and compare several prompts.
Cross-check with other tools or reliable resources.
Basics of exchange traceability.
Security, confidentiality, and compliance
Do not share sensitive, personal, or confidential information.
Organization’s internal rules (GDPR, intellectual property, confidentiality).
Best practices for professional use:
Data anonymization
Human review before publication
Retention and deletion of conversations.
Ethics and responsibility
Understanding biases (social, cultural, linguistic).
Promoting inclusive, neutral, and responsible use.
Transparency in AI usage: stating when a generative tool has been used.
Environmental and societal impact of AI.
