Announcement on “Considerations when using Generative AI in the Chuo University Curriculum” (For Students)

Summary Version(214KB)
Japanese version

1.Introduction

Generative AI is becoming increasingly widespread as a convenient tool for everyday use. It also assists in university lectures, enhancing your learning efficiency and aiding deeper thinking. However, generative AI is not a panacea; its use carries specific risks and responsibilities. Students should fully understand the following properties and precautions, master generative AI on a daily basis, and use it for learning with initiative.

2.Fundamental Properties of Generative AI

Generative AI produces the most plausible response to a user's prompt. That is, the output is determined probabilistically. This probability calculation utilises the vast amount of data from which the generative AI has learned.
Consequently, generative AI generally possesses the following properties:

• - Data is learned and information leakage is possible: Information input by users into generative AI services may be used for training the AI, depending on settings. That is, input information is stored by the services. If confidential information is input into generative AI services, it may be reflected in responses provided to other users.
• - There is a possibility of hallucinations being generated: Due to the nature of responses being generated probabilistically, the generated content may contain errors. Information that appears plausible but is incorrect is termed a “hallucination.” When using generative AI, it is necessary to verify that the responses do not contain hallucinations.

3.Engaging with Generative AI in Daily Life

Gemini, Copilot, ChatGPT, and Claude are representative generative AI services, but beyond these, AI is utilised in various contexts: Rufus provided by online retailer Amazon within its site; Grok on social media platform X and Meta AI on Facebook; translation application DeepL; Notion AI offered by workspace tool Notion; AI integrated into various PDF readers, and even auto-summary features within email applications like Gmail, summary displays for search results in internet browsers, and applications for editing/generating images, videos, and audio. AI is utilised in numerous contexts (*1). While the convenience of these tools and features is widely recognised, you may unknowingly be inputting information into AI and using its outputs.
Therefore, it is essential to remain vigilant about the following points in your daily interactions:

• - Exercise caution when inputting information: It is now impossible to predict where inputted information may be reused. You must refrain from inputting confidential or personal information. While attention often focuses on text-based input, opportunities to use voice input or self-taken images for input have increased recently. Even such audio or images could potentially be used for training generative AI.
• - Exercise caution with automatically generated information: Even if you have not input any information yourself, certain generative AI services, tools, or application features may automatically output AI-generated content (regardless of type, such as text, audio, images, and videos). Be mindful that information displayed on your smartphone or computer may have been generated by AI. Always be aware that visible information could potentially contain hallucinations or ethical issues.

In this AI era, where caution regarding AI usage is increasingly emphasised across various contexts, it is worth noting that similar warnings have existed since the early days of the internet's expansion – such as “do not input personal information online” and “always verify information found through internet searches.” Now in the AI era, we may be using AI unconsciously. Therefore, whenever using smartphones or computers, keep these precautions firmly in mind.

(*1) as of February 2026

4.Engaging with Generative AI in the Classroom

Generative AI holds the potential to enhance your learning. However, you must not become overly reliant on its use to the point of abandoning your own critical thinking. In the AI era, it becomes even more crucial for each individual to take ownership of their learning. Strive to use it in ways that foster your critical thinking and analytical skills.

4.1 Examples of Appropriate Generative AI Use in Class

In class, generative AI can be used for the following:

• - Generating ideas
  Example:Organising multiple perspectives on an idea, expanding upon an idea
• - Collecting information
  Example: Efficiently gathering information on a topic of interest (However, it is vital to fact-check whether the information is reliable)
• - Drafting and refining text
  Example: Obtaining perspectives differing from one’s own ideas, gaining hints to improve one's own writing, checking for typos and errors
• - Using as a learning partner
  Example: Using it for managing learning or project progress, having it serves as a reflection partner
• - Utilising for self-directed learning
  Example: Repeatedly studying challenging topics, having English compositions corrected

However, even in courses that do not utilise generative AI, it is expected that the experience gained in such courses will lead to appropriate AI utilisation. For instance, they should provide opportunities to develop methods for identifying reliable sources from the perspective of your academic discipline, as well as the knowledge, skills, and attitudes necessary for critically evaluating information. Through your various regular coursework, cultivate the ability to appropriately utilise generative AI.

4.2 Points to Note

When utilising generative AI, adherence to “Academic Integrity” (*2) is required. “Academic Integrity” refers to an attitude that consistently upholds the following six fundamental values:

• - Honesty: Being truthful in all academic activities and communication, free from dishonesty or deception.
• - Trust: Building open relationships and environments where trust in each other's honesty enables secure exchange of ideas and collaboration.
• - Fairness: Learning through one's own efforts, avoiding unfair conduct, within evaluations and treatment that are fair and consistent, where standards and practices are reasonable and transparent.
• - Respect: Valuing oneself, others, diverse opinions, and the learning process itself; participating proactively and making the most of opportunities.
• - Responsibility: Taking responsibility for one's actions, adhering to norms and policies, and maintaining an unwavering stance against misconduct or undesirable influences from others.
• - Courage: Maintaining the resolve to do what is right, even in difficult situations, by upholding one's values and integrity, speaking out against wrongdoing, and standing firm.

These values translate into concrete guidance on “how one should act,” forming the foundation for translating ideals into actual behaviour within the university, in classes, and in daily life. Please note that inappropriate use of generative AI may result in academic sanctions or disciplinary action.
When upholding academic integrity, be mindful of the following:

Essential Practices
When utilising generative AI, it is imperative to clearly distinguish between your own thoughts and assertions and information derived from the output generated. To achieve this, adhere to the following guidelines:

• - Explain in your own words: When submitting assignments, it may be good to use generative AI. However, "just change the output result from bullet points to plain text" or "just change the end of the word" does not reflect your own thoughts and words. To ensure “responsible use of generative AI,” to create deliverables that you can take responsibility for in your own words.
• - Fact-check: Verify the basis of generated information using reliable literature or databases.
• - Maintain records: Similar to citing existing literature, university regulations stipulate that when using generative AI outputs in reports, you must explicitly state how and which parts were used, including the following details:

• - Name and version of the generative AI system used
• - Date of use of the generative AI system

Additionally, keep records so that the following information can be provided if required:

• - Information on prompts and environment settings input into the generative AI system
• - Output content from the generative AI system

Points to Note
When utilising generative AI, keep the following in mind:

• - Consider ethical perspectives: Ensure generated content does not infringe human rights, contain discriminatory expressions, or reflect biases.
• - Interpret critically: AI may incorporate biases derived from its training data. Evaluate outputs objectively and critically.
• - Stay updated with new information: Generative AI evolves daily, so make a habit of gathering the latest insights.
• - Beware of over-reliance: Excessive dependence on generative AI risks diminishing your own critical thinking skills. Remember, consulting teachers, peers, or checking books in the library may yield faster and more accurate information than asking AI. Utilise diverse university resources alongside generative AI.

Prohibited Actions
The following actions, which contravene academic integrity, are strictly forbidden.

• - Direct copying: Copying and pasting AI responses verbatim and submitting them as reports or assignments is prohibited by the university.
• - Infringement of Rights: To avoid infringing copyrighted works or intellectual property, do not input others' papers or works into generative AI without the rights holder's permission.
• - Inputting Personal or Confidential Information: Inputting your own or others' names, addresses, non-public personal information, or unpublished research or activity details into generative AI can lead to serious information leaks.

Points to Confirm
The faculty, department, or program to which you belong, or the teacher in charge of each course you enrol in, may establish special rules governing the use of generative AI. Adhere to these instructions.

• - Example Rule 1:

“In this course, generative AI may be utilised for gathering, organising, and summarising information; generating ideas; and drafting outlines. However, submitting content generated by AI verbatim as reports or presentation materials is not permitted.”
→ You may use generative AI within the specified circumstances.

• - Example Rule 2:

“Concerned that its use may prevent the attainment of learning objectives, this course prohibits the use of generative AI in all contexts.”
→ Generative AI must not be used for any learning activities within this course.

Consulting
If you are uncertain whether using generative AI contravenes academic integrity or have questions about the rules for using generative AI, you must consult the teacher in charge.

Be sure to check the precautions and rules for using generative AI at Chuo University at Announcement on “Considerations when using Generative AI in the Chuo University Curriculum

(*2) International Center for Academic Integrity [ICAI]. (2021). The Fundamental Values of Academic Integrity. (3rd ed.). www.academicintegrity.org/the-fundamental-valuesof-academic-integrity (Visited on 10 February 2026).

5.When Using Generative AI Tools at Chuo University

Chuo University provides free access to the paid versions of Google's Gemini (using with your university-issued email account) and Microsoft's Copilot Chat (using with your university-issued Microsoft account – modify your email account name by replacing "@g" with "@m"), for which the university holds contracts.

When logged in with your university account, information entered into these services is generally not used for training purposes, allowing for relatively secure usage. As a Chuo University student using generative AI, you should endeavour to use these tools exclusively while logged in with your university account.

However, there are no “absolutes” in any matter. Adhere strictly to the general precautions for using generative AI and refrain from entering personal or confidential information.

Generative AI Services Available for Use (University-Contracted Services)

	Provided via Google Workspace Gemini	Provided via Google Workspace NotebookLM	Provided via Microsoft 365 Copilot Chat
Service Content	Chat-based support for text creation, answering complex questions, code generation, brainstorming, etc.	Chat-based support for information retrieval, summarization, and relationship analysis based on uploaded materials.	Chat-based support for text creation, answering complex questions, code generation, brainstorming, etc.
How to Use * Please refer to the respective Help Centers for detailed operating instructions. ■ NotebookLM Help Center https://support.google.com/notebooklm/ ■ Gemini App Help Center https://support.google.com/gemini/	Sign in to Google Workspace using your university-wide email account (@g.chuo-u.ac.jp).	Sign in to Google Workspace using your university-wide email account (@g.chuo-u.ac.jp).	Sign in to Microsoft 365 using your university-issued Microsoft account (@m.chuo-u.ac.jp).
Risks of External Leakage and Model Training	Prompts and uploaded materials (usage data) are not used for AI model training. However, since data is stored in the cloud as history, account compromise (e.g., via cyberattacks) could lead to information leaks.	(Same as Gemini)	(Same as Gemini)
Handling of Personal and Confidential Information	Exercise extreme caution when handling personal, private, or confidential information.	Due to the risks mentioned above, avoid unnecessary input (prompts or file attachments) in principle.	Do not use information obtained from part-time jobs or internships without permission.

Generative AI Services Requiring Caution (Individually Used Services)

	Free Generative AI Services (No cost)	Individually Contracted Services (Free or Paid)
Risks of External Leakage and Model Training	These services can often be used without registration. Some do not allow "opt-out" (disabling the use of input for training), raising concerns that your data may be used for model training or future AI responses.	Terms and conditions (pricing, training settings, commercial use, etc.) vary by service. Most require user registration and login.
Precautions and Terms of Use	Even if the service is free, there may be restrictions such as a prohibition on commercial use. You must check the individual terms of service.	You must review and agree to the terms of use regarding data training controls and commercial use before use.
Handling of Personal and Confidential Information	Do not use information related to personal privacy or any information that would be problematic if made public.	It is strongly advised to avoid entering personal privacy or confidential information.

Table: Created by Chuo University Information Systems & Security Office