By Xi Lin
As artificial intelligence becomes more common in education, one important question emerges: Can AI provide feedback as effectively as humans, especially in teacher preparation?
A recent study explores this question by comparing AI-generated feedback (ChatGPT) with human peer feedback in a literacy methods course for preservice teachers. The findings offer timely insights into how future educators learn, reflect, and improve their teaching practice in an AI-supported world.
Why Feedback Matters in Teacher Education
Feedback plays a central role in helping preservice teachers improve their lesson planning, critical thinking, and instructional decision-making. Traditionally, peer review has been widely used to support:
- Reflective thinking
- Collaborative learning
- Pedagogical development
With the rise of AI tools like ChatGPT, feedback is no longer limited to human interaction. AI can now provide instant, rubric-based suggestions, raising an important question:
“Does faster feedback mean better learning?”
The Study: AI vs. Human Peer Review
The study used a quasi-experimental design with 25 preservice teachers:
- ๐ฅ Human feedback group (n = 9): Students reviewed each other’s lesson plans in class
- ๐ค AI feedback group (n = 16): Students used ChatGPT to generate feedback
All students:
- Created a guided reading lesson plan
- Received feedback (human or AI)
- Revised their work
- Reflected on the feedback experience
Data included:
- Survey results (critical thinking and peer learning)
- Open-ended responses
- Written reflections
Key Findings: What Did We Learn?
1. Both AI and Humans Support Critical Thinking
Students in both groups reported improved critical thinking.
AI helped by:
- Providing structured, rubric-aligned feedback
- Identifying gaps quickly
- Supporting revision efficiency
This suggests that AI can function as a useful cognitive scaffold, helping students refine their work.
2. Only Human Feedback Fostered Peer Learning
Here is where the difference becomes clear:
- Human feedback → significantly improved peer learning
- AI feedback → no significant effect on peer learning
Why? Students emphasized that human interaction provides:
- Dialogue and discussion
- Shared understanding
- Immediate clarification
- Emotional support
In contrast, AI lacks true interaction and collaboration, which are essential for peer learning.
3. Strengths and Weaknesses of AI Feedback
Strengths of AI:
- ⏱️ Immediate and always available
- ๐ Strong alignment with rubrics
- ๐งพ Detailed and structured suggestions
Limitations of AI:
- ❗ Sometimes irrelevant or inaccurate
- ๐ค Lacks contextual understanding
- ๐ฌ Feels impersonal or “robotic”
- ๐ Can be inconsistent across responses
Some students even questioned its reliability and authenticity.
4. Human Feedback Brings What AI Cannot
Human peer feedback stood out for:
- ❤️ Emotional support and encouragement
- ๐ฏ Contextual relevance (real classroom understanding)
- ๐ค Trust and collaboration
- ๐ง Pedagogical nuance
Students reported feeling:
- More confident
- More engaged
- More supported
These socio-emotional and relational aspects are still beyond AI’s current capabilities.
The Future: A Hybrid Feedback Model
Rather than choosing between AI and humans, the study suggests a more effective approach:
Combine both!
AI for:
- Fast, structured, rubric-based feedback
- Identifying surface-level issues
Humans for:
- Deep discussion and reflection
- Emotional and contextual support
- Collaborative meaning-making
A hybrid feedback model may provide the best of both worlds.
Implications for Educators
For teacher educators and instructional designers:
- ✔ Use AI as a support tool, not a replacement
- ✔ Teach students how to critically evaluate AI feedback
- ✔ Design activities that include both AI and human interaction
- ✔ Emphasize AI literacy and ethical use
Final Thoughts
AI is transforming education—but not in a way that replaces human connection.
This study shows that while AI can
enhance efficiency and support critical thinking,
Human interaction remains essential for meaningful learning.
๐ The future of
education is not AI or humans.
๐ It is AI and humans—working together.
Reference
Yang-Heim, G. Y. A., & Lin, X. (2026). Preservice teachers’ perceptions of AI-and human-generated feedback on lesson plans. Cogent Education, 13(1), 2624898. https://doi.org/10.1080/2331186X.2026.2624898
