Critical thinking for AI

How to maintain critical judgment when using AI?

  • ChatGPT
  • Claude
  • Microsoft Copilot

1960 € per session, excl. VAT

In short

For professionals who analyze and make decisions, this training teaches how to evaluate AI responses. Participants learn to verify evidence and distinguish robust analyses, producing a verification grid and a sourced recommendation note. This contributes to more prudent and well-founded decision-making.

Why this training?

The problem
Professional teams struggle to evaluate the reliability and validity of information generated by artificial intelligence.
What it costs today
Decisions based on fragile analyses or unverified information can be made, leading to errors or rework.
Why this skill
This skill is necessary for prudent and better-founded decision-making when integrating ChatGPT, Claude, and Microsoft Copilot input into analysis and recommendations.
What you gain
Participants and the company gain autonomy in evaluating AI responses and improve the reliability of recommendations issued.

Objectives of this training

  • Distinguish facts, opinions, inferences, and causality in AI outputs.
  • Trace back to sources, dates, scopes, and methods used by AI.
  • Test the validity of figures, citations, and information provided by AI.
  • Request alternatives, objections, and invalidation conditions for AI proposals.
  • Compare two AI-generated responses without confusing consensus with truth.
  • Present a conclusion with its evidence, limitations, and the decision reserved for humans.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Data analysts interpreting AI outputs
  • Project managers integrating AI tools into workflows
  • Decision-makers evaluating AI-based recommendations
  • Report writers using AI for information synthesis
  • Consultants advising on AI solution adoption

Prior skills

  • AI Fundamentals and AI Prompting completed, or equivalent experience (a learning prerequisite, not a registration condition).
  • Ability to formulate clear queries for AI tools
  • Understanding of business context and decision-making challenges
  • Knowledge of ethical principles related to data use
  • Practice of critical analysis of various information
  • A paid subscription to an AI solution offering the features used, the organisation’s authorisation for third-party tools, and the actual accounts and access rights.

To prepare before the session

  • One computer per participant.
  • Access to the selected AI tools (accounts and licenses verified before the session).
  • A business case to work on, using fictional or anonymized data.

What you will work on

  1. Recognizing Reasoning Shortcuts

    The participant can identify cognitive biases in AI responses.

  2. Verifying Evidence

    The participant can assess the soundness of evidence and information provided by AI.

  3. Encouraging Contradiction

    The participant can use AI to generate contradictory perspectives and evaluate alternatives.

  4. Formulating a Responsible Judgment

    The participant can elaborate an informed recommendation integrating AI contributions with discernment.

What you take away

  • A verification grid for AI responses.
  • A contradictory analysis of AI proposals.
  • A sourced and reasoned recommendation note.
  • Practical supports and completed templates.

Method and assessment

  • Demonstrations with explanations using a concrete case
  • Guided exercises followed by application to your role
  • Peer review of results and source verification
  • An individual or team action plan

Assessment is formative: participants complete and review a practical case linked to the intended deliverable (A verification checklist, a counter-analysis and a sourced recommendation memo). Criteria include meeting the brief, accuracy, checks performed and the ability to explain decisions. Each participant identifies an application to test in their own work.

Frequently asked questions

What concrete outcome can my team expect after this training?

Your team will be able to analyze AI tool responses with more discernment. They will be capable of producing more robust recommendations and identifying gaps in generated information, which will reduce the risks of errors in projects.

Who should participate in this training and what knowledge is expected beforehand?

This training is for professionals who analyze information, write recommendations, or make important decisions. Participants should already know how to formulate queries for AI tools and understand the challenges related to data use in their business context.

What types of data or documents will be used and how will confidentiality be protected?

We will work on concrete business cases using fictional or anonymized data to preserve confidentiality. Tools like ChatGPT, Claude, and Microsoft Copilot will be explored without compromising the security of your organization's sensitive information.

What kind of exercise will be performed on my business case during the session?

You will apply the learned methods to a business case you bring. This will include evaluating AI responses, building a verification grid, and drafting a reasoned recommendation note, identifying evidence and limitations.

What are the limitations of tools like ChatGPT, Claude, Microsoft Copilot, and who is responsible for the final validation of the work?

AI tools have limitations in terms of source reliability, contextual understanding, and the ability to generate original analyses. The training emphasizes that final validation and decision-making always rest with humans. AI is an aid, not a substitute for professional judgment.

How does the training adapt to the specificities of our business or our internal use cases?

The training is designed to be applied to your own business cases. Practical exercises are performed on examples you choose, allowing direct adaptation of skills to your organization's realities. The methods taught are generalizable to various professional contexts.

Last updated: