Given the very rapid development of AI tools, these guidelines may evolve over time. It is essential that students stay informed about these developments. 

Given the growing integration of artificial intelligence (AI) tools into academic and scientific practices, particularly generative AI tools such as ChatGPT and Copilot, we need to think carefully about how it is used and make sure we comply with the highest standards of intellectual and research integrity.

INRS guidelines for the responsible use of AI in graduate studies

Best practices in the use of artificial intelligence

Challenges and limitations of AI

INRS guidelines for the responsible use of AI in graduate studies

INRS’s Graduate Studies and Student Success Service has developed guidelines on artificial intelligence. They are set out in the document “Guidelines for the responsible use of artificial intelligence in graduate studies” and are summarized below.

Summary

Prioritize your academic learning: Use AI as a supportive tool—it should not replace the development of essential academic skills such as critical thinking, independent research, and writing.

Obtain prior authorization: Always check whether AI tools are allowed in a given context. Follow institutional policies and specific guidelines to avoid inappropriate use.

Take responsibility for AI-generated content: You are fully responsible for any content produced with AI. Always verify the accuracy, relevance, and originality of the information before including it in your work.

Understand associated risks: Be aware of potential risks, including:

  • Privacy: Sensitive or personal data may be exposed.
  • Plagiarism: AI-generated content may include unoriginal material without clear attribution.
  • Cybersecurity: Some AI tools may put your information at risk of cyberattacks.
  • Algorithmic bias: AI outputs may reflect biases present in the training data.
  • Environmental impact: Consider the energy footprint of AI tools and use them responsibly.

Maintain a critical mindset and practice transparency when using AI tools.

CONSULT THE GUIDE (pdf)
French version (PDF)

Basic concepts

Artificial intelligence (AI)

“Artificial intelligence (AI) refers to the series of techniques which allow a machine to simulate human learning, namely to learn, predict, make decisions and perceive its surroundings. In the case of a computing system, artificial intelligence is applied to digital data.”

Source : Déclaration de Montréal, Glossaire, https://declarationmontreal-iaresponsable.com/glossaire

Generative AI (GAI)

“Generative AI refers to all artificial intelligence techniques used to produce content via algorithms and big data, usually in the form of text, sound, video, or image files. Generative artificial intelligence can lead to plagiarism or misinformation, particularly when data has been deepfaked. More broadly, chatbots  are types of generative artificial intelligence.”

Source (traduction de) : Office québécois de la langue française, Grand dictionnaire terminologique, https://vitrinelinguistique.oqlf.gouv.qc.ca/fiche-gdt/fiche/26561649/intelligence-artificielle-generative

“The term ‘generative’ refers to AI’s capacity to automatically create information based on large volumes of data on which it has been trained. AI does not simply copy and paste what it has analyzed; rather, it imitates, refines, and produces entirely new outputs, generated through a statistical recomposition of patterns and structures identified during training. These outputs, commonly referred to as content, may take the form of text, images, music, or computer code.” 

Source : Université de Genève, Generative Artificial Intelligence: A Guide for the Academic Community, Introduction to generative AI  https://www.unige.ch/numerique/digital-university/ia-generative-guide-unige/introduction-ia-generative  

Differentiating between AI and GAI

Large language models (LLMs)

“An LLM is considered ‘large’ because it contains a vast number of parameters—on the order of billions—each representing a piece of information. It is a ‘model’ because it consists of a neural network trained on extensive text data to perform a range of non-specific tasks. It is ‘language-based’ because it replicates the syntax and semantics of human natural language by predicting the most probable continuation for a given input. This also enables it to possess a form of general ‘knowledge’ derived from the training texts.” 

Content reproduced with the permission of the University of Geneva
Source : Intelligence artificielle générative. Guide pratique de l’UNIGE. Introduction à l’intelligence artificielle générative. Université de Genève.

Challenges and limitations

When using AI tools, be aware of the associated challenges and limitations and consider whether they are being used appropriately and responsibly.

A hindrance to acquiring essential skills: AI should be a complementary tool and not a substitute for intellectual work. It should not compromise the acquisition of essential academic skills such as critical analysis or writing, research, or problem-solving skills.


Exposure of sensitive data: Personal, confidential, or unpublished data such as research results should never be shared with AI tools without explicit approval. These tools may store or analyze data in unpredictable ways, which can pose serious risks to confidentiality and cybersecurity. Remember—all uses involving sensitive data must comply with the INRS Privacy Policy.


Cybersecurity risk: Sharing data via online AI platforms may expose users to cyberattacks or data breaches.

Algorithmic bias and reliability of responses: AI tools do not guarantee that the responses they generate are accurate or objective. The results may also reflect biases in the training databases, which can hinder diversity and inclusiveness.


Intellectual property and plagiarism: Using AI-generated content may lead to potential breaches of intellectual property rights such as copyright. AI tools may rely on protected data or works without the permission to do so, exposing users to litigation. AI tools must be used in accordance with the rules governing citations, references, and attribution of sources.


Environmental and human issues: AI tools have direct and indirect impacts on the environment (e.g., raw materials to manufacture AI equipment, water to cool servers and energy to train AI tools, etc.) but also on society, particularly the labour market (i.e., automation of certain tasks, new skills for employees, etc.). etc.).

Resources and references

See all

Ressources

Sources consulted for the guidelines

Some AI glossaries (non-exhaustive list)

Additional resources and tools

Intellectual Integrity
Intellectual Integrity
Learn how to cite your sources
Learn how to cite your sources
Student Services and Resources (GARE)
Student Services and Resources (GARE)

Questions? Contact us at the following address

pedagogie.sesre@inrs.ca