동서대학교 주메뉴
전체메뉴
생성형 AI 활용 가이드라인
Guidelines for Using Generative AI
The purpose of these guidelines is to help the faculty, students, and other members of Dongseo University respond wisely to the spread of Generative AI and to use this tool effectively in learning. These guidelines are to be voluntarily observed.
What is Generative AI?
Generative AI refers to a technology that generates new data using artificial neural networks. It understands user intent through prompts, learns from the given data, and uses what it has learned to create new content such as text, images, audio, and video.
[Common] Ethical Use of Generative AI
1. AI-generated content should always be reinterpreted or supplemented with additional thoughts for productive use
Critically assess the content of AI-generated outputs (e.g., assignments) and ensure to verify their truthfulness.
Effort should be made to reinterpret and modify AI responses to create new and original outputs.
2. When using AI-generated content, check for any biases or discriminatory elements.
It is important to verify that the results generated by AI do not contain biased or discriminatory content related to gender, age, disability, region, race, religion, nationality, etc. If such content is found, it must be removed, and efforts should be made to prevent negative impacts on society as a whole.
As the use of Generative AI becomes more widespread, it is essential to respect human dignity and use AI-generated content in a way that pursues the common good, ensuring that all members of society can lead dignified lives.
3. AI-generated content should be checked for copyright infringement or plagiarism.
The use of Generative AI allows for easier access to more information, which increases the possibility of unintentional copyright infringement or plagiarism.
Users must be aware that they could face legal consequences if AI-generated outputs cause harm to others, infringe on their rights, or affect safety. The responsibility for such issues lies solely with the user.
4. Sensitive information (e.g., confidential data, personal information) should not be entered into AI systems.
When a prompt is entered into Generative AI, it generates outputs in the form of text, images, etc. The information provided by the user is not deleted but stored on servers. The AI uses this stored information for learning and relearning, which can lead to the generation of different outputs.
Users should exercise special caution to avoid inputting sensitive personal information or confidential information related to their organization into prompts.
Guidelines for Faculty on the Use of Generative AI
1. Faculty should decide whether and how to use AI in their courses, taking into account the specific nature of the course.
Before integrating Generative AI into their teaching, faculty should understand how it works, recognize its limitations, and make informed decisions based on the course's characteristics.
Faculty can explore ways to actively incorporate Generative AI to support our university's educational goal of nurturing globally creative individuals who embody the "The Only One" value.
2. Faculty should clearly outline the use of Generative AI in their course syllabi.
If Generative AI will be used in a course, this should be explicitly stated in the syllabus, and all relevant details must be clearly communicated to students.
Faculty can collaborate with students to establish and share guidelines that include any limitations, scope of use, methods, and precautions related to the use of Generative AI.
3. Faculty should emphasize and consistently reinforce ethical standards in the use of Generative AI.
Before using Generative AI, faculty must provide students with pre-education on ethical standards, including ethics, responsibility, truthfulness, and bias.
Faculty should guide students on the ethical standards and minimum rules they should consider when using Generative AI, and these should be explained consistently throughout the course.
4. Faculty should clearly explain the precautions, plagiarism issues, and academic dishonesty related to the use of Generative AI.
Faculty should thoroughly explain the advantages, disadvantages, and precautions of using Generative AI, ensuring students fully understand them.
Faculty must stress the importance of properly citing sources when using AI-generated content.
Faculty should make it clear that using AI-generated outputs without modification can constitute plagiarism or academic dishonesty.
Guidelines for Students on the Use of Generative AI
1. In courses involving Generative AI, it is important to fully understand the instructor’s guidelines.
The guidelines outlined in the course syllabus regarding the use of Generative AI, as well as any additional instructions provided by the instructor, should be carefully followed.
Responsibility for any issues that arise from submitting AI-generated content as part of graded assignments lies with the student, especially if the content is used without proper modification.
2. Ethical and responsible use of Generative AI is crucial.
Before utilizing Generative AI, it is essential to understand the ethical guidelines provided by the instructor or as outlined in the AI usage guide.
Given that Generative AI can produce biased or inaccurate information, it is important to carefully check the content for bias and accuracy before use.
3. Verification of the accuracy of AI-generated content is necessary.
AI-generated content should not be used without first being critically evaluated.
Since Generative AI can produce false information or include references to existing works, careful review and verification of the content’s accuracy and originality are required.
4. Proper citation of sources when using AI-generated content is mandatory.
When Generative AI is used to create content, it is necessary to clearly state the manner of AI usage and properly cite all sources.
Direct use of AI-generated content without modification may constitute plagiarism or academic dishonesty, and students bear the responsibility for such actions.
Examples : (Reference) Chung-Ang University’s guidelines on the use of Generative AI
|
Text-based Generative AI |
ChatGPT4(2024. 06. 01). “Prompt content.” Generated or written using OpenAI’s ChatGPT 3.5. https://chat.openai.com/ |
|
Image-based Generative AI |
Stable Diffusion(2024. 06. 01). “Prompt content.” Generated or created using Stable Diffusion online. https://stablediffusionweb.com/ |
Checklist for Responsible Use of Generative AI
|
Have AI-generated results been properly labeled as content created with Generative AI? |
YES□ |
NO□ |
|
Have texts, audio, images, or other materials that might infringe on the rights of others been avoided when using Generative AI? |
YES□ |
NO□ |
|
When entering questions or information into Generative AI, have steps been taken to ensure that no content defames or discriminates against specific individuals? |
YES□ |
NO□ |
|
Is the content generated by Generative AI free from hate speech directed at individuals, organizations, or specific groups, and does it avoid expressing biased opinions or judgments? |
YES□ |
NO□ |
|
Has sensitive information, such as personal data or corporate secrets, been protected by not providing it to Generative AI when obtaining information or creating content? |
YES□ |
NO□ |
|
Misinformation and Disinformation Has care been taken not to deliberately input inaccurate or manipulated information into Generative AI for the purpose of creating fake news, spam, or other misleading content? |
YES□ |
NO□ |
|
Has the AI-generated content been checked to ensure that it is free from biased perspectives? |
YES□ |
NO□ |
|
Have steps been taken to critically evaluate and cross-check the information provided by Generative AI, recognizing that it may not always be accurate? |
YES□ |
NO□ |
|
Before relying on Generative AI, has there been sufficient independent thought and consideration, with AI being used only as a supplementary tool? |
YES□ |
NO□ |
|
Instead of using AI-generated content as-is, has it been reinterpreted or enhanced with personal thoughts and ideas for productive use? |
YES□ |
NO□ |
<Source: Korea Communications Commission & National Information Society Agency (2023). "Ethical Guidelines for Generative AI," p.69>