Examples and Student Perceptions in AI-Augmented Technical Writing
Content created and copyrighted by Clint Guymon in 2025 using reveal.js
Brigham Young University
Remember
Understand
Apply
Analyze
Evaluate
Create
Bloom's Taxonomy
Creating
Writing Technical Content is Creative Thinking
AI or Large Language Models (LLMs) can help
Students
- organizing thoughts
- reducing leaps in logic
- getting helpful feedback
AI or Large Language Models (LLMs) can help
Instructors
- drafting templated initial feedback
- generating rubrics
Topics Not Discussed Today:
- How to use AI/LLM tools
- Ethical and responsible use of AI/LLM in technical writing
- Abuse of AI/LLM tools that may negatively impact critical thinking skills
First Example
0.5 Credit Lab Class with a Multi-week Project
Students had to submit a group technical report and were allowed to use AI tools to help them draft and improve their writing
Initial Instructor Feedback can be Automated and then Individually Tailored
Example Code to Work with the API
What do students think of this?
74 Students participated in the survey
A majority agreed LLM's can be helpful to learning
A minority agreed LLM use by the instructor can be helpful to their learning
Second Example
2 Credit Class on Computational Tools with a Multi-week Project
Students had to submit a group technical report and were allowed to use AI tools to help them draft and improve their writing
The same approach used as in the First Example to provide feedback:
- AI used to draft templated initial feedback
- Individually tailored feedback for each report
The feedback format is slightly different; it's clearer what the AI feedback vs the instructor feedback was. What do students think?
45 Students participated in the survey
A majority agreed LLM's can be helpful to learning
A minority agreed LLM use by the instructor can be helpful to their learning
Transparency with AI Use
Students may be more likely to view instructors use of AI as helpful to their learning when AI use is clearly outlined:
| Set | Positive | Negative |
|---|---|---|
| No Transparency | 27 (36%) | 47 |
| Transparency | 20 (44%) | 25 |
| p-value is 0.25 from Fisher-Exact Test | ||
Faculty Survey
A majority agreed LLM's can be helpful to student learning
A minority agreed LLM use by the instructor can be helpful to their learning
Evidence that:
- Students perceive that they are sufficiently instructed in AI use even with little to no instruction
- Students are more likely to see the value of AI in their learning than the instructor's use of AI
- Many instructors view student use of AI/LLM as positive
- Many instructors are unsure of AI/LLM use for themselves in providing feedback on student writing
- Use of AI by the instructor for giving feedback on technical writing can save time by providing a second opinion within a structure
Recommendations
- AI can help with facilitating learning and improving technical writing
- Students need to be taught how to properly use AI
- zero shot, single shot, few shot, chain of thought, etc.
- Instructors need to be aware of how it can be used and abused
- Transparency in AI use is critical