Last updated on: April 21, 2023
AI technology is constantly evolving. The content on this page will be monitored and updated as new information becomes available.
Generative artificial intelligence (AI) is a category of technology that can produce new content, including audio, code, images, text, simulations, and videos. Recent breakthroughs in the field, including the public release of ChatGPT in November 2022, are already radically changing the way individuals and organizations approach content creation. Awesome Generative AI provides an extensive list of generative AI technologies, categorized by type, including text, code, image, video, and audio generators. You are encouraged to check the privacy policies and terms of use for any generative AI tools before applying them.
There are several AI technologies currently garnering significant attention in higher education, such as OpenAI’s ChatGPT, DALL·E 2, and Whisper.
As with any source of information, you need to be critical of generative AI and its trustworthiness. Understanding the way that AI generates outputs, and thinking carefully about the limitations of these technologies, can show you why it is important to be cautious when using them.
Generative AI applications are not primary sources of information and will give you responses whether or not they are appropriate or correct. They seek out information they feel is relevant based on algorithms, without being concerned about ethics, or accuracy. Generative AI tools do not process information the way you do; they just ‘make up’ what they think is a correct series of words to fill in gaps based on statistical likelihoods.
These tools are ‘trained’ on large amounts of data from the internet and databases. They consume the data without differentiating between factual and fake information. Most of the content on which many large language models, like ChatGPT, were trained are written in English and reflect Western norms and perspectives. As the use of generative AI increases, the output may reinforce and amplify the biases of dominant cultures, leading to responses that may ignore, or silence, critical perspectives on a topic.
Generative AI tools like ChatGPT also use the data you input to train their machines. That data is neither confidential nor secure. Anything tools like ChatGPT tell you (and anything you tell them) will be kept and used by the owners of artificial intelligence generators, therefore, be mindful of handing over information the tool can store and use in other contexts. The servers that host generative AI models could also be vulnerable to data breaches or hacking, which could expose user data. If the tools have not been internally vetted by YU and TFS, we cannot be assured that they have appropriate privacy and security protocols in place.
Generative AI accesses content from all over the internet and there are some concerns about whether these technologies are stealing intellectual property and plagiarising. Most of the content that is generated by AI is gathered from online sources, without crediting those sources. For example, there are concerns from artists that DALL·E 2 has been reproducing artistic works without permission as it draws from and reproduces derivatives of artists’ work, and some companies have prohibited the sale of AI-generated images.
Uploading information such as assignment descriptions to artificial intelligence platforms might also mean you breach copyright yourself inadvertently.
There are significant financial barriers to accessing many generative AI systems, particularly as they are increasingly privatized and commodified. Most require a transaction fee or subscription, or operate on a "freemium" model, enabling limited free access to the tool while only allowing paid subscribers to access the full feature set. For example, OpenAI, the creators of ChatGPT, which was released to the public for free during a research preview, launched ChatGPT Plus on February 1, 2023, a monthly subscription plan that promises users faster response times and priority access to new features and functionalities.
There are many potential applications of generative AI, and it is how these tools are used that determines if it is appropriate. Though AI can be used to support your studies, there are also times when it is not appropriate to use. In particular, submitting an assessment that is not your own work is a breach academic integrity.
What Might Appropriate Use of Artificial Intelligence Look Like?
If you have been asked to use generative AI in one of your assessments, it would be acceptable to use the tools with appropriate acknowledgement. See How to cite AI for more information.
Generative AI can help you make practice quiz and exam questions. You can use these to prepare for assessments. You could also use generative AI to brainstorm or create a summary of a topic for you, and you can critically evaluate if this is correct, based on your notes from the unit. Remember though, answers given by generative AI are not necessarily correct, or accurate. Always confirm answers with course content.
You may be using AI generated material as an example in your work. You will need to make certain you are acknowledging this. You should also read the information above about the reliability of AI before using it as a source. If you are using generated images, audio files and/or codes, check the copyright details for the generator you are using and reference appropriately.
You may be allowed to use AI to improve your writing, depending on the rules for completing your assignment. Some AI algorithms, like Quillbot and Grammarly, are designed to help you enhance your written expression. It’s important to remember that, depending on the nature and goal of the assignment, this type of use may not be acceptable. Always defer to the instructions shared by your faculty and/or your program.
What Might Inappropriate Use of Generative Artificial Intelligence Look Like, and How Might It Impact Your Studies?
It’s important to remember that all work you submit for assessment must be your own, so check you are not misrepresenting your own work and capabilities. Assessments are how you demonstrate your understanding of unit concepts and learning outcomes and will help you develop skills for your future. Where the use of AI is permitted, make sure you are using it in a way that is responsible and ethical.
If your faculty, program, and/or assignment description have stated that you are not permitted to use AI for a particular assessment, but you use it anyway, you are engaging in academic misconduct.
Just like any source, if you are using AI generated material as part of your assessment, to inform your argument, or as an example, it needs to be acknowledged in-text and in your reference list, or through a declaration. If you don’t do this, it might be considered academic misconduct.
Assessments help you to consolidate your understanding, and demonstrate the skills needed to perform in professional roles after graduation. To successfully complete a program, you must independently demonstrate achievement of course and program learning outcomes.
While AI tools may be used to explore disciplines, spark ideas, and support studying, you should always acknowledge and document the use of generative AI in any coursework (see How to Cite AI). Assignments and tests may have explicit requirements or restrictions on the use of generated AI material. By submitting coursework, you assert that you have respected these requirements. Submitting and representing AI generated material as your own work is considered an academic offense as outlined in section 5.6.2 of the academic calendar:
[…] all forms of cheating and academic fraud and misrepresentation, not only those listed here, will be dealt with according to the policy and procedures outlined below.
| Academic Integrity Breach | Definition from the Academic Calendar (5.6.2 Academic Offenses) | The risk of using generative artificial intelligence |
|---|---|---|
|
Plagiarism |
Plagiarism is the act of representing someone else’s work as one’s own |
If you use artificial intelligence to create the content of your assessments for you, you are claiming this work as your own. This is a misrepresentation of your ability and knowledge. Artificial intelligence generators collate other existing ideas and present them in their own words without appropriately attributing them. You may inadvertently plagiarise someone because of the way generative artificial intelligence is presenting the information to you, and, text based submissions may be flagged through Turnitin. |
|
Cheating |
Anything done to dishonestly or unfairly gain unearned academic advantage, grades, or credits is a serious breach of academic integrity. |
When you use artificial intelligence to create assessment content for you, you are asking a third party (the developer of the artificial intelligence) to create your assessment for you. Even if you don’t pay to use the AI tool, this may be a form of contract cheating. If the use of AI is not authorized, it may also be deemed as a means of unfairly gaining academic advantage. |
It is important that the work that you submit is your own work. Under the Student Conduct Policy (Section 5.6 of the Academic Calendar), all work that you submit claiming to be your own, must be your own work. An academic breach may apply if you are submitting work that is not your own and claiming that it is, including if it has been developed by AI.
Like any tool, artificial intelligence can be used appropriately, and inappropriately. You can reflect on how you use these tools by asking yourself (a few questions):
Always check with your program and faculty first! Before starting your assignment, confirm with your instructor that you are permitted to use generative AI tools such as ChatGPT. Unauthorized use of generative software may constitute an academic integrity offence.
Remember to include a citation whenever you incorporate content generated by AI in your assignment. For example, if you paraphrased or quoted text generated by ChatGPT, cite the source in text and in your references list. Neglecting to cite this material may constitute an academic integrity offence. Since your reader can't access the exact AI-generated text, you may also consider including a copy of the text as an appendix, or as part of supplemental materials. See How to cite AI for more information.
As part of using AI responsibly and ethically, it is important to be transparent and acknowledge how you have used AI in your work. If you used AI to generate ideas or an outline for your assignment, include a statement in your work indicating which AI tools you used and how you used them. Neglecting to include this statement may constitute an academic integrity offence.
Be aware that there is a possibility that the content AI generates is biased, outdated, or false. Always evaluate the information generated by AI tools. Students are responsible for the accuracy, factuality, and truthfulness of AI content used to assist with, and in, assessments.
Faculty regularly monitor student work for significant changes in the style or quality of their submissions or content and references that, while plausible, are factually incorrect or do not exist. In addition to their own efforts, an AI Indicator is also integrated within Turnitin, our approved, enterprise-level plagiarism detection tool. For more information on the AI Indicator, review this summary of its detection capabilities.
If the course instructor has concerns regarding student work and believes that an academic offense may have been committed, they will follow the procedure outlined in section 5.6.3 of the Academic Calendar. As noted in Section 5.6.6, students have the right to appeal policy decisions where there are appropriate grounds for doing so.
Standards for citation of AI are evolving. What follows are suggestions based on current APA guidelines. Your program may have specific citation requirements. Always check with your program first.
AI generated content may not be considered as an acceptable source for your course work. Be sure to evaluate the content carefully and check with your instructor if you are permitted to use it as a source. If you’re using ChatGPT and other AI tools for coursework, include a description of how you used AI within your work. For example, you can describe in your introduction how you used ChatGPT to generate an outline for your presentation. When citing AI-generated text, remember to include both in-text and references list citations. For more information, see APA Style’s How to cite ChatGPT.
Developer. (Version Year). Name of generative AI tool (Version number) [Type of AI Model]. URL of tool
OpenAI. (2023). ChatGPT (Mar 23 version) [Large language model]. https://chat.openai.com
Remember to include the prompt you used whenever you quote or paraphrase text generated by an AI tool. Since your reader can’t access the exact AI-generated text, you may also consider including a copy of the text as an appendix or as part of supplemental materials. If you are including a copy of the generated text, you may refer to it in the body of your text, or as part of the in-text citation such as in the example below.
Given the prompt “What are large language models?” the text generated by ChatGPT described them as “artificial intelligence systems that have been trained on vast amounts of text data, with the goal of understanding natural language and generating human-like responses to text-based inputs” (OpenAI, 2023).
When provided with a follow up prompt of “What are examples of large language models?” ChatGPT identified OpenAI’s GPT-3, NVIDIA’s Megatron, and Google’s LaMDA as some of the examples of large language models (OpenAI, 2023; see Appendix A for the full transcript).
AI-generated content may not be considered as an acceptable source for your course work. Be sure to evaluate the content carefully and check with your instructor if you are permitted to use it as a source.
If you are inserting the AI-generated image in your coursework, start with a figure number followed by an image title. Beneath the image, include a note with details about the prompt and the source (see example below). Remember to include the full citation in your References List.
Developer’s Name. (Version Year). Name of AI tool (Version number) [Type of AI model]. URL of tool
OpenAI. (2023). DALL-E (Version 2) [Large language model]. https://labs.openai.com
Though generative AI may make assessments seem easier, using these tools to do all the work for you will be detrimental to your learning and career. There is also no guarantee that any content generated by AI is reliable, true, correct, or of sufficient standard to pass an assessment task. If you need assistance with your assessments, or any part of your studies, there are lots of places you can go for support. If you have questions about or would like a refresher on academic integrity, there are also a number of resources that you can access, including:
Understand the different types of information, and help you choose the best information for your assignment.
List of Library Staff at YU provides advice for how to direct a reference question.
