🌐Overview

AI is changing not only how we create content, but how we understand authorship, expertise, labor, and accountability. Its adoption brings up new questions about ownership, data ethics, and transparency. It also highlights the need for strong systems to support its responsible use. As AI tools become more advanced, we need better frameworks. These frameworks should ensure fair, sustainable use while recognizing their limits.

📖 Table of Contents

Establish Clear Citation and Disclosure Standards
- Citation Guidelines
Copyright
Privacy and Data Security
- Best Practices for Project Leads on AI and Data Security
Sustainability Is About More Than Reuse
- Ecological Sustainability
- Labor Sustainability
- Pedagogical Sustainability
- Institutional Sustainability

✔️ Establish Clear Citation and Disclosure Standards

Transparency about AI use is essential to maintain academic credibility and trust in your OER. Attribution Statements Should Include:

The tool used (e.g., “Portions of this section were generated using ChatGPT 4.0.”)
A brief description of how it was used (e.g., “for outlining” or “to generate initial quiz questions”)
The fact that all AI content was reviewed and revised by human authors
The US Copyright Office also requires that "if a work contains more than a de minimis amount of AI-generated material, the applicant should disclose that information and provide a brief statement describing the human author’s contribution."

📌 Citation Guidelines

American Psychological Association. Released April 7, 2023

Format: Author. (Date). Name of tool (Version of tool) [Large language model]. URL
Example: OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat
In-Text Citation Example: (OpenAI, 2023)

Chicago Manual of Style. Released in spring 2023

Format: 1. Author, Title, Publisher, Date, url for the tool.
Examples
- Including information about the prompt: 1. ChatGPT, response to "Explain how to make pizza dough from common household ingredients," OpenAI, March 7, 2023, https://chat.openai.com/chat.
- If information about the prompt has been included within the text of your paper: 1. Text generated by ChatGPT, OpenAI, March 7, 2023, https://chat.openai.com/chat.

Modern Language Association. Released March 17, 2023

Format: "Description of chat" prompt. Name of AI tool, version of AI tool, Company, Date of chat, URL.
Example: "Examples of harm reduction initiatives" prompt. ChatGPT, 23 Mar. version, OpenAI, 4 Mar. 2023, chat.openai.com/chat.
In-Text Citation Example: ("Examples of harm reduction")

Jump back to top

✔️ Copyright

AI-generated materials live in a gray area of ownership. The AI company does not own the output of AI, nor does the prompt engineer own it. If you copy text from AI directly, you do NOT own the copyright and cannot put a CC license on it. The details are still being worked out; for more information, review the Copyright and Artificial Intelligence Report generated by the Copyright Office. You should consider some key quotes from the report when using AI in a work.

Copyright does not extend to purely AI-generated material or material with insufficient human control over the expressive elements.
Prompts alone don’t give users enough control to be considered the authors of AI-generated content. They are just instructions, not protected creative input. While future AI tools might provide users with more control, AI still does too much creative work for now.
Using AI to help (not replace) human creativity doesn’t change the question of whether the final work can be protected by copyright.

Two cases exemplify a potential line for originality.

In the “Théâtre D’opéra Spatial” decision, the U.S. Copyright Office denied registration of a work because it "contained more than a de minimis amount of content generated by AI" and because the contributions of the author and the AI were “inextricably merged, inseparable contributions”. Allen argued that “the underlying AI-generated work merely constitutes raw material,” which he “transformed through his artistic contributions” (a series of at least 624 prompts, adjusting the image in Adobe Photoshop, and upscaling the image with Gigapixel AI).
The Copyright Office approved registration for “A Single Piece of American Cheese” based on “Selection, coordination, and arrangement of material generated by artificial intelligence.” Invoke outlines an extensive series of human contributions to support their claim.

💡Tip: The lines between AI being a "tool to assist" and a "stand-in" for human creativity are still fuzzy. This information is evolving, but what seems to be important is that AI use is disclosed and that the "creative work" is human-generated.

Jump back to top

✔️ Privacy and Data Security

Data security has become a critical concern as AI tools become more integrated into academic work and Open Educational Resources (OER) development. Many AI platforms gather user data. This includes sensitive student info, unpublished research, and copyrighted content. If not correctly managed, this data can be stored, reused, or exposed without the user’s knowledge or consent. Institutions must ensure AI tools follow privacy laws like FERPA. They should also uphold ethical standards for academic integrity and protect intellectual property.

Faculty and project leaders should not enter identifiable student data or proprietary content into public AI tools. They should use platforms that have clear data policies and secure storage practices instead. Recommended: Your institution’s IT and FERPA compliance offices should review your AI policy before implementation.

📌Best Practices for Project Leads on AI and Data Security

Use Institutionally Approved Tools. Choose AI platforms vetted by your institution’s IT or legal team to ensure they meet data privacy standards (e.g., FERPA, HIPAA, where applicable).
Avoid Inputting Sensitive Data. Never enter student names, ID numbers, grades, unpublished research, or licensed materials into public AI tools.
Read the Fine Print. Review the AI tool’s privacy policy and terms of use. Avoid platforms that claim ownership of user inputs or allow third-party data sharing.
Create Clear Use Guidelines. Develop and share internal policies for how your team should use AI tools, including dos and don’ts for data entry, review, and storage.
Use Local or Private AI Models Where Possible. For sensitive work, consider using locally hosted or institutionally licensed AI models that don’t transmit data externally.
Secure Storage of AI-Generated Outputs. Store drafts and final outputs on secure, institution-approved platforms (e.g., SharePoint, institutional Google Drive) rather than personal devices.
Document AI Use Transparently. To ensure transparency and accountability, keep records of what AI tools were used, for what purpose, and by whom, especially for OER creation.
Provide Team Training. Ensure that faculty and collaborators understand data risks and how to responsibly use AI tools in line with institutional and ethical guidelines.

Jump back to top

✔️ Sustainability Is About More Than Reuse

Sustainability in OER is traditionally framed around updates, maintenance, and long-term availability. But AI brings additional dimensions that demand attention:

📌 Ecological Sustainability

Generative AI requires massive computational resources. A single query to a large model like GPT-4 can consume more energy than a search engine request. Water cools data centers, which raises concerns about environmental justice. This issue is especially important in drought-prone areas like California. Project leads should

Prioritize low-bandwidth tools or on-device models where possible
Offer non-AI alternatives to students concerned about the ecological impact
Include environmental sustainability in AI tool selection criteria

📌 Labor Sustainability

AI can save time on first drafts. However, it also adds extra work. You still need to review, edit, and validate sources. Plus, you must ensure accessibility and compliance. Faculty labor shifts from creation to critical oversight, often invisibly. Without explicitly recognizing these evolving labor demands, we risk burnout, inequity, and undercompensation. Sustainable practices should include

Transparent crediting of AI support in contributor agreements
Clear delineation of roles between AI, authors, editors, and reviewers
Capacity-building for those unfamiliar with AI workflows

📌 Pedagogical Sustainability

Using generative AI tools to create quizzes, feedback, or lessons can lead to generic results. Students might get responses that lack context or reflect only dominant cultural norms. In the long term, this can erode our learning materials' educational depth and cultural responsiveness. To sustain pedagogical integrity, OER projects must

Embed culturally relevant pedagogy in prompt design
Align AI outputs with course learning outcomes—not just surface comprehension
Regularly audit AI-generated material for accuracy, tone, and student-centered relevance

📌 Institutional Sustainability

As we move forward, our decisions about transparency, authorship, accessibility, and equity will shape the norms and expectations for educational content creation in the years to come. Colleges must develop systems-level planning that ensures AI supports, rather than undermines, our public mission. Emerging questions:

What training do faculty and OER contributors need to use AI responsibly?
Who maintains and monitors AI-influenced OER content over time?
How do we track the evolution of an AI-generated resource and ensure it stays up-to-date and accurate?
Can we build open-source AI tools that reflect community values, rather than relying exclusively on corporate APIs?

Jump back to top