In each passing day of reading online content, an ongoing question in the back of my mind is, “was this written with Generative AI?” Being able to confidently detect synthetic text is debatable, thus we are seeing much interest in how content creators are being transparent in stating their GenAI use.
Recognizing the expanding use of GenAI, and restricting it being not only unfeasible, but perhaps unfair, for the 2025 Open Education Awards for Excellence we added to the nomination form an optional field for “Generative AI Transparency” (see a document version of the entire form):
In a spirit of transparency, we ask all nominators to indicate how Generative AI was used to submit this form. e.g. to organize statements, to assist for writing in a language that is not your primary one, etc. Please make sure all written statements are reviewed and composed as best as possible from a human perspective. This has no bearing on the review of your nomination.
As I reviewed all of the nominations as they arrive in our system, last year I could see there was a notable use of GenAI, much of it for helping nominators who’s first language is not English, but also uses for fitting responses to word limits, as well as refining wording to better match the form’s requirements.
We also noted affirmations of human editing and control of the final text. As well a good number of nominations affirmed they did not use GenAI. One nominator declared that no GenAI was used in their nomination, yet the first line in the description included the rephrasing reply one gets from systems like ChatGPT.
This was one human’s summary, but I was curious what we might learn from summarizing with GenAI. Form responses were exported along with the country represented in each nomination (but no identifying personal details) into a Google spreadsheet then loaded into NotebookLLM for analysis.
Overall Summary of GenAI Transparency
The first summarization was:
Out of 199 nominations, 183 or 92% responded to this question.
This does not mean 92% use if GenAI as the question was open ended, and this included the “no” type responses. The automated summary by NotebookLLM was:
The provided excerpts track the self-reported transparency of Generative AI usage across a variety of nomination submissions originating from diverse geographic locations, including Canada, Mexico, the United States, and various countries in Europe and Africa. A significant portion of respondents confirmed using tools like ChatGPT, generally employing them for linguistic refinement and organizational tasks. This assistance frequently involved using AI to polish grammar, improve clarity, or translate content from a native language, such as Spanish, into English. Other functional uses included having the AI organize complex ideas, structure the narrative, or shorten text to meet specific character limitations. Across the submissions that utilized AI, the consensus was that the generative tools served only as an assistant, emphasizing that the core ideas and final content were reviewed and approved by human authors.
This seems fairly accurate from my own reading of the responses to the transparency question.
GenAI Tool Usage
I also wanted to see from the open ended responses if there was a means to indicate what tools were used. The prompt I entered was:
Prompt: Create a chart of the Gen AI tools mentioned and the number of times mentioned.
which produced a chart and some other explanatory notes. It seems not surprising for the time that ChatGPT was the most common tool listed. Note as well that we did not ask nominees to specify the GenAI tool used- this reflects what they offered to share.
| Generative AI Tool | Mentions |
| ChatGPT / Chat GPT | 36 |
| Grammarly | 4 |
| Microsoft Copilot | 3 |
| DeepL | 2 |
| AI Perplexity / Perplexity | 1 |
| DeepSeek | 1 |
| Google Gemini | 1 |
| LLMs (Large Language Models) | 1 |
| Notebook LM | 1 |
Notes on Mentions:
- ChatGPT was frequently mentioned in conjunction with the general term “Generative AI” (e.g., “Generative AI (ChatGPT)”) but was counted here as a specific tool . It was also used for tasks ranging from translation to content reduction and organization .
- Grammarly was cited specifically for its role in checking spelling, flow, and grammatical errors .
- Microsoft Copilot was utilized for generating APA formatted citations and reviewing the nomination in the final editing stages .
- DeepL was used to support the translation of single terms or sentences from German into English .
- The term LLMs (Large Language Models) was used once as a category of tools to compile information and clean up text .
No Use of GenAI
I next asked NotebookLLM to characterize the statements about not using GenAI with the question:
Prompt: How many total nominations explicitly reported no generative AI use?
The results indicate 31% of all the responses to this question were stating no GenAI use, which we can thus say 69% were transparent in their use.
Based on the source provided, 57 total nominations explicitly reported no generative AI (GenAI) use during the submission process.
These explicit statements include phrases such as:
- “No generative AI was used to submit this form.”
- “Generative AI was not used to complete this nomination.”
- “No AI used.”
- “Gen AI was not used for this submission.”
- “Generative AI was not used in any way to develop, organize, improve, or otherwise support this nomination. It was written and revised wholly by human hands.”
- “I have not used AI to write this statement…”
- “This application did not use artificial intelligence at all.”
- “Aucune utilisation d’IA générative pour ce formulaire.”
- “none used”
Categories of GenAI Use
Lastly I wanted to ask NotebookLLM to characterize the stated uses with the prompt:
List the significant types of GenAI use, with 1-3 sentence description. Include under each up to 3 quotes that represent this use.
The full summary is detailed below.
The utilization of Generative AI (GenAI) during the nomination submission process fell into several significant categories, primarily focused on refining language, organizing content, and managing technical constraints like translation and character limits.
1. Linguistic Refinement, Grammar Correction, and Polish
Nominees frequently used GenAI as a sophisticated writing assistant to enhance the quality of their prose, correct grammatical issues, and ensure textual coherence. This involved refining word choice, improving sentence flow, and polishing the final text for a professional presentation.
- “Generative AI was used for linguistic refinement of the text and in order to guarantee that it was free of grammatical errors.”
- “Generative AI (ChatGPT) was used to help organize and draft written statements for this nomination, including structuring responses and refining language for clarity and impact.”
- “AI was used to refine sentence structures and improve grammar usage” .
2. Drafting, Structuring, and Content Organization
A core application of GenAI was managing the structural complexity of the submission form. Nominees utilized tools to organize initial ideas, draft sections of text, outline responses, and synthesize input from multiple human contributors to create a unified and well-structured final document.
- “Generative AI (ChatGPT) was used to help organize and draft written statements for this nomination, including structuring responses and refining language for clarity and impact.”
- “Generative AI was used to organize and synthesize statements that were made by multiple human contributors to this nomination form to unify the voice and avoid duplication.”
- “I used generative AI tools to help organize my ideas and draft portions of the nomination text, especially to ensure clarity and polish.”
3. Translation and Non-Native English Support
For many international nominees, GenAI served as a critical tool to overcome language barriers, specifically when English was not their native language. It was used to accurately translate responses drafted in languages like Spanish, Hindi, or Chinese into English, or to assist with wording and phrasing to improve fluency.
- “I used the support of an AI tool to help me draft and express these ideas in English, as it is not my native language.”
- “I originally wrote all my responses in Hindi… I used Generative AI solely for accurate and faithful translation from Hindi to English.”
- “Generative AI was used to assist in organizing and refining written statements in English, which is not the nominee’s primary language.”
4. Content Reduction and Character Management
Nominees frequently used GenAI to summarize longer texts or modify existing content to adhere to specific spatial constraints within the form. This included asking the AI to shorten the original text or count characters to ensure the submission met required word or character limits for each section.
- “AI was used to shorten the original nomination to fit within the character limit.”
- “I only used ChatGPT to help shorten my text to meet the character limits.”
- “AI was used to reduce word count to make sure the opening section met the character limit.”
Changes for the 2026 Nomination Form
For the 2026 OEAwards nominations, still in progress, I adjusted the question, first by asking just a yes/no question:
Did you make use of any Generative AI tools in the developing of this nomination?
At this point, 60% of the responses are yes.
If a nominator checked yes, the followup question is:
Please share specifically what Generative AI tools were used and for what purpose(s). This should be in the form of declarative statements like “ChatGPT was used to reduce my narrative to fit the character limits” or “Claude provided suggestions for alignment with the principles and goals of open education”
In this way I am hoping we get very specific details on what was used and how. I was strongly influenced here by the work of Amanda Grey and colleagues at Kwantlen Polytechnic University on their AI Declaration Statement template for OERs, especially on the structure of the statements (for more see Amanda’s demo from Open Education Week).
From my scanning of this year’s nominations, the use cases summarized above seem to cover what is being shared, but the declaration statements provide more insight into how GenAI has been used for the 2026 OEAwards, leading to more understanding then “AI or Not.”
You can expect a followup analysis of the GenAI transparency statements for 2026, but what we are really looking forward to is sharing all of the nominees and the unfolding of the award process.