Skip to Main Content

Soil Science

About AI Generated Text

Before using ChatGPT for assignments, refer to UBC's Academic Integrity page and be sure to check with your instructor to find out if AI tools are permitted. 

ChatGPT (Generative Pre-trained Transformer) uses natural language processing techniques to respond to user-generated prompts. Put simply: You pose a question or provide a prompt, and the tool replies using natural language. 

Is ChatGPT trustworthy?
Responses can reflect biases from the people who wrote the original text used in the training dataset. ChatGPT was trained using a massive dataset of text written by people and pulled from the Internet. ChatGPT is not connected to the Internet and the data used to train it was collected prior to 2021. Information may contain errors and be outdated.

Fake citations
It has been well noted that ChatGPT and similar AI tools generate false citations. ChatGPT makes stuff up!

Privacy
OpenAI (the company that designed ChatGPT) collects data from ChatGPT users. Their privacy policy states that this data can be shared with third party vendors, law enforcement, affiliates, and other users. Input into ChatGPT cannot be deleted. Why does this matter? If you ask ChatGPT about sensitive or controversial topics, that data cannot be removed or deleted and others may have access to it. 


Some content on this page has been adapted or adopted from the open licensed An Instructor's Guide to Teaching & Learning with Technology @UNBC by UNBC CTLT 

APA

In a post from April 7, 2023, APA offers some guidance on citing AI-generated materials and AI software.

 

Text Created by AI tools

First, they suggest ensuring you describe how you used the AI tool in your research in a method section or comparable section of your paper.

APA also suggests that given that AI generated content like chats are not created by a person, that they cannot be considered personal communication. Instead, treat the content as an algorithm's output, and credit the author of the algorithm with a reference list entry and in-text citation.

You can also put the full text of long responses in an appendix or online supplemental materials.

Reference List

OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat

In-Text Citation

For a direct quote or paraphrase, use the following format. Note: OpenAI is the author not ChatGPT.

(Author, Year)

Example:

(OpenAI, 2023)

 

References to ChatGPT or other AI models

APA suggests using the template for software to cite ChatGPT or other AI software.

Reference

OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat

In-Text Citation

Parenthetical citation: (OpenAI, 2023)

Narrative citation: OpenAI (2023)

Vancouver Style

Reference List

Although there is no specific guidance offered by Vancouver Style editors, AI content is considered personal communication and/or non-recoverable data, and therefore should not be included as a reference in your reference list according to the Vancouver Style.

In-Text Citation

AI-generated content can be cited in-text using the following format:

Type of Communication, Communicator, Date (DMY)

You can also include further information, such as the prompt you offered the AI in type of communicaiton.

Examples:

In an online chat with Open AI's Chat GPT (23 March 2023)...

A chat generated a potentially useful script for dealing with these interactions (ChatGPT response, prompt for "How to resolve conflict with a coworker", 24 March 2023).

...(Grammerly paraphrase, personal communication, 22 February 2023).

IEEE

IEEE does not cite AI generated text for publication and does not consider AI generated text a valid reference. 

CSE

Reference List

Although there is no specific guidance offered by CSE, AI content is considered personal communication and/or non-recoverable data, and therefore should not be included as a reference in your reference list according to CSE style.

In-Text Citation

AI generated content can be cited in-text with the following format of other unpublished materials

(Author, Format, Date [DMY]).

(OpenAI, Chat GPT, 23 March 2023)

AMS (American Mathematical Society)

Reference List

Although there is no specific guidance offered by the American Mathematical Society, AI generated content is generally considered personal communication and/or non-recoverable data. According to AMS style, you should be including this in your reference list. AMS reference lists are organized alphabetically, with each reference assigned a number.

Use the following format:

Author Name, Format Type, Year

Examples:

[5] Open AI Chat GPT chat, 2023

[5] Open AI Chat GPT chat response to prompt "How to resolve conflict with a coworker," 2023

[4] Grammarly paraphrase, 2023

In-Text Citation

AI-generated content can be cited in-text using using the reference's assigned number

Examples:

In a chat with Open AI’s Chat GPT [5], the author discovered...

American Society for Microbiology (ASM)

Reference List

Although there is no specific guidance offered by American Society of Microbiology, AI content is considered personal communication and/or non-recoverable data, and therefore should not be included as a reference in your reference list according to ASM style.

In-Text Citation

AI generated content can be cited in-text with the following format of other unpublished materials

(Author's first initial(s). Author's last name, Affiliated institution/organization/conference, Location [if relevant], Date [DMY]).

(OpenAI’s Chat GPT 23 March 2023)

Evaluating for AI Generated Content

There are some detection tools that can help determine whether or not text has been generated by a human, AI or both. Here are some examples:

All sources you find on the internet should be evaluated carefully whether they are AI generated or not. Using the SIFT method is a helpful way to determine if the source you have found is reliable and expert information. 

 

SIFT graphic

"SIFT (The Four Moves)" by Mike Caulfield, re-used and adapted under CC BY 4.  

 

There are four moves to help you evaluate information you find on a website. Using the Library search, Summon, helps find research articles, books and more to help verify information.

 

 

  1. Stop. Do you know the website or where the information is coming from? Can you verify the claims in the website? Before you click "share" make sure you have verified the information.
  2. Investigate the Source. Who is providing the information? What is their level of expertise? Do they have an agenda that might influence what is said and how the information is presented? Using the Library search, Summon, can help verify the information and provide multiple reliable sources.
  3. Find Better Coverage. Look into multiple sources that can help verify the claim. What are the experts saying about the topic? Where else has the story been covered?
  4. Trace claims, quotes, and media back to the original context. Trace the claim you find on social media, or a news clip, or a website you land on back to the source to get the context for the claim. 

 

AIs and Authorship

Several publishers have offered guidance on how AI tools should be cited or used in writing:

Nature and other Springer Journals advise that Large Language Models (LLMs) like ChatGPT will not be accepted as credited authors on research papers. However, researchers using LLM tools should document their use in the methods or acknowledgement section (https://www.nature.com/articles/d41586-023-00191-1, January 24, 2023)

arXiv’s policy on ChatGPT and similar tools states that authors are required to report any significant use of sophisticated tools in a manner consistent with subject standards for methodology, and that generative AI tools should not be listed as an author. Authors are also responsible for all content of their articles, including that generated by AI language tools, and should be mindful that AI generated content may include inappropriate language, plagiarized content, errors, mistakes, incorrect references or misleading content. (https://blog.arxiv.org/2023/01/31/arxiv-announces-new-policy-on-chatgpt-and-similar-tools/, February 7, 2023)

JAMA also states that nonhuman artificial intelligence, language models, machine learning or similar technologies to not qualify for authorship, but should be reported in the Acknowledgements or Methods section. Likewise, authors take responsibility for the integrity of that content. (https://jamanetwork.com/journals/jama/fullarticle/2801170, January 31, 2023)

Science has created an Artificial Intelligence (AI) policy which states: “Text generated from AI, machine learning, or similar algorithmic tools cannot be used in papers published in Science journals, nor can the accompanying figures, images, or graphics be the products of such tools, without explicit permission from the editors. In addition, an AI program cannot be an author of a Science journal paper. A violation of this policy constitutes scientific misconduct.” (Science URL as of March 23, 2023)

Taylor & Francis states that AI tools cannot be listed as an author, and their use must be acknowledge and documented appropriately (https://newsroom.taylorandfrancisgroup.com/taylor-francis-clarifies-the-responsible-use-of-ai-tools-in-academic-content-creation/, as of March 23, 2023)

Citation Guides

American Mathematical Society (AMS)

Reference List

Although there is no specific guidance offered by the American Mathematical Society, AI generated content is generally considered personal communication and/or non-recoverable data. According to AMS style, you should be including this in your reference list. AMS reference lists are organized alphabetically, with each reference assigned a number.

Use the following format:

Author Name, Format Type, Year

Examples:

[5] Open AI Chat GPT chat, 2023

[5] Open AI Chat GPT chat response to prompt "How to resolve conflict with a coworker," 2023

[4] Grammarly paraphrase, 2023

In-Text Citation

AI-generated content can be cited in-text using using the reference's assigned number

Examples:

In a chat with Open AI’s Chat GPT [5], the author discovered...

 

American Society for Microbiology (ASM)

Reference List

Although there is no specific guidance offered by American Society of Microbiology, AI content is considered personal communication and/or non-recoverable data, and therefore should not be included as a reference in your reference list according to ASM style.

In-Text Citation

AI generated content can be cited in-text with the following format of other unpublished materials

(Author's first initial(s). Author's last name, Affiliated institution/organization/conference, Location [if relevant], Date [DMY]).

(OpenAI’s Chat GPT 23 March 2023)

 

APA

In a post from April 7, 2023, APA offers some guidance on citing AI-generated materials and AI software.

Text Created by AI tools

First, they suggest ensuring you describe how you used the AI tool in your research in a method section or comparable section of your paper.

APA also suggests that given that AI generated content like chats are not created by a person, that they cannot be considered personal communication. Instead, treat the content as an algorithm's output, and credit the author of the algorithm with a reference list entry and in-text citation.

You can also put the full text of long responses in an appendix or online supplemental materials.

Reference List

OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat

In-Text Citation

For a direct quote or paraphrase, use the following format. Note: OpenAI is the author not ChatGPT.

(Author, Year)

Example:

(OpenAI, 2023)

 

References to ChatGPT or other AI models

APA suggests using the template for software to cite ChatGPT or other AI software.

Reference

OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat

In-Text Citation

Parenthetical citation: (OpenAI, 2023)

Narrative citation: OpenAI (2023)

 

CSE

Reference List

Although there is no specific guidance offered by CSE, AI content is considered personal communication and/or non-recoverable data, and therefore should not be included as a reference in your reference list according to CSE style.

In-Text Citation

AI generated content can be cited in-text with the following format of other unpublished materials

(Author, Format, Date [DMY]).

(OpenAI, Chat GPT, 23 March 2023)

 

IEEE

IEEE does not cite AI generated text for publication and does not consider AI generated text a valid reference. 

 

Vancouver Style

Reference List

Although there is no specific guidance offered by Vancouver Style editors, AI content is considered personal communication and/or non-recoverable data, and therefore should not be included as a reference in your reference list according to the Vancouver Style.

In-Text Citation

AI-generated content can be cited in-text using the following format:

Type of Communication, Communicator, Date (DMY)

You can also include further information, such as the prompt you offered the AI in type of communicaiton.

Examples:

In an online chat with Open AI's Chat GPT (23 March 2023)...

A chat generated a potentially useful script for dealing with these interactions (ChatGPT response, prompt for "How to resolve conflict with a coworker", 24 March 2023).

...(Grammerly paraphrase, personal communication, 22 February 2023).

AI Resources at UBC