Choosing the Right AI Tool for Academic Writing: thesify vs. ChatGPT
Written by Alessandra Giugliano, MSc, University of Amsterdam
thesify vs. ChatGPT: Testing AI Tools for Academic Writing Success
Tempted to use ChatGPT for your next academic essay? You’re not alone. Generative AI’s widespread popularity and instant text-generation capabilities make it an appealing choice for students. However, relying on ChatGPT could lead to serious issues, from breaching academic integrity to receiving feedback that’s too generic to make a meaningful difference.
Last week, we tested thesify’s ability to provide actionable and ethical feedback on an undergraduate sociology essay I wrote years ago. The essay analyzed how gender is “done” in a social gathering setting, and thesify impressed me by identifying gaps in my analysis, clarifying my thesis, and guiding me to strengthen my arguments—all without generating any text. You can read more about that test in last week’s blog post: Testing thesify: How This AI Tool Saved My Undergrad Paper.
This week, we decided to put ChatGPT to the same test using the exact same essay. Our goal? To see how its feedback compares to thesify’s and whether it truly supports academic writing in an ethical and effective way. The results revealed clear differences between the two tools and their approach to helping students and researchers improve their work. This blog dives into the key areas of comparison, so if you’re deciding between thesify and ChatGPT, keep reading to find out which tool truly delivers.
ChatGPT vs. thesify: Which AI Tool Helps Without Cheating?
In the age of AI, academic integrity is a nuanced issue, with educators and students alike navigating what is considered ethical use of these tools. During testing, ChatGPT frequently generated text—even after being explicitly prompted not to—raising concerns about its alignment with academic standards. On the other hand, thesify stood out for its design, which prioritizes ethical, learning-focused feedback that helps writers improve without overstepping boundaries.
thesify’s Approach:
Promoting Critical Thinking: thesify’s feedback consisted of targeted questions that encouraged deeper reflection and analysis, ensuring the user maintained full control over their work.
Preserving the Writer’s Voice: suggestions respected the essay’s context and tone, avoiding intrusive or pre-written content that might undermine originality.
Learning-Oriented Feedback: Its guidance was strictly aimed at improving the writer’s skills, balancing constructive criticism with validation to preserve the essay’s original voice.
ChatGPT’s Approach:
Over-reliance on Text Generation: Even when instructed to avoid creating content, ChatGPT repeatedly offered pre-written examples that risked crossing the line into academic misconduct.
Limited Contextual Awareness: While ChatGPT provided surface-level suggestions, its reliance on text generation often detached its feedback from the essay’s academic objectives, making it less actionable.
Disconnected Solutions: ChatGPT’s feedback often prioritized convenience over meaningful academic growth or encouraging novel ideas.
Key Difference: thesify’s ethical design ensures students improve their essays without risking violations of academic standards, while ChatGPT’s reliance on content generation highlights critical flaws for those seeking to maintain integrity in academic writing.
For more on crafting ethically sound academic work, check out 9 Tips for Using AI for Academic Writing (Without Cheating).
Thesify vs. ChatGPT: Which AI Provides Better Thesis Feedback?
When tasked with analyzing the thesis of the test sociology paper, the differences between thesify and ChatGPT were immediately apparent. ChatGPT initially assumed the wrong sentence in the essay was the thesis, and when corrected, it polished the wording but failed to address alignment or weaknesses.
In contrast, thesify went deeper by identifying where the thesis lacked clarity and connection to the assignment’s goals, such as addressing gender binaries and inequality. This feedback not only improved the thesis but also strengthened the essay as a whole.
thesify’s Feedback:
Identifying Gaps: thesify flagged that the thesis didn’t fully align with the assignment’s focus and failed key academic tests like “So What?” and “How and Why?”
Actionable Guidance: Instead of rewriting, thesify offered clear suggestions to refine the thesis and tie it to broader sociological concepts such as gender inequality and performativity.
Strengthening Connections: It pointed out where sections of the essay failed to connect to the thesis, guiding improvements in coherence and argument structure.
For more insights on crafting a strong thesis, check out The Basics of Thesis Writing: How to Develop a Strong Thesis Statement.
ChatGPT’s Feedback:
Misidentifying the Thesis: ChatGPT assumed an incorrect sentence was the thesis, leading to irrelevant feedback.
Superficial Edits: Once corrected, it focused on stylistic polishing but ignored deeper issues like alignment and purpose.
Lack of Depth: ChatGPT didn’t provide insight into why the thesis was weak or how to strengthen its connection to the essay’s arguments.
Key Difference:
thesify provided ethical, detailed feedback that helped refine the thesis while improving the essay’s overall coherence and academic rigor. It focused on clarity, alignment, and actionable steps, empowering learning and improvement. ChatGPT, by contrast, relied on quick fixes, ignoring the deeper academic purpose and leaving significant gaps unaddressed.
For a deeper dive into thesify’s teaching approach, read Rewriting the Story: How Feedback Turns Good Essays Into Great Ones.
Does Your Essay Align with the Assignment? How thesify Excels Where ChatGPT Falls Short
Meeting the specific goals of an assignment is just as important as producing well-written content in an academic essay. For more on writing with intention, check out Writing With Purpose: Making Every Word Count. To test this, we provided both thesify and ChatGPT with the same essay, along with detailed assignment instructions. While thesify delivered precise, actionable feedback rooted in the assignment’s goals, ChatGPT struggled with surface-level evaluations and failed to meaningfully assess whether the essay fully met its purpose.
thesify’s Purpose-Driven Feedback
Built Around Assignment Details: thesify uses the information entered—such as the assignment type and instructions—to assess how well the essay meets its purpose.
Example from the Test Essay: thesify identified that while the observations in the test paper were strong, they lacked critical connections to systemic inequality and broader sociological frameworks, a key requirement of the assignment.
Criteria-Based Breakdown: Feedback was organized by assignment requirements, with each goal marked as "met" or "not met."
Example from the Test Essay: The test essay succeeded in integrating course readings but fell short in comparing observations to another social category, leaving a major criterion unmet.
Overall Score: thesify assigned a comprehensive "Needs Improvement" score, clearly showing how the essay performed against the assignment’s overall purpose.
Specific, Actionable Advice: Detailed guidance included practical steps, such as linking observations to theorists like Lorber and Zimmerman, ensuring the essay fully addressed the assignment’s expectations.
ChatGPT’s Weaknesses:
Misleading Assessment:
ChatGPT claimed the essay “achieved its purpose effectively,” despite significant gaps that thesify identified, such as the lack of comparison to another social category. This overgeneralized evaluation left critical areas unaddressed.
Surface-Level Feedback:
ChatGPT provided a list of observations, such as how costumes challenged gender norms, but it failed to assess deeper alignment with the assignment’s purpose. It didn’t highlight whether specific criteria were met or not.
No Overall Score:
Unlike thesify, ChatGPT didn’t assign a comprehensive score for how well the essay met its purpose. This omission left gaps in understanding the paper’s overall performance.
Broad, Vague Suggestions:
While ChatGPT suggested exploring intersectionality further, it didn’t tie this recommendation back to the assignment’s core goals or provide actionable steps for improvement.
Limited Use of Assignment Details:
Despite being prompted with the same instructions, ChatGPT’s feedback was generic and disconnected from the specifics of the task. It failed to analyze whether the essay met the detailed criteria outlined in the prompt.
Key Difference:
thesify’s structured, goal-oriented feedback ensured the essay fully responded to the assignment’s expectations, while ChatGPT’s surface-level approach risked misleading writers about their work’s quality.
Strengthening Academic Arguments: thesify vs. ChatGPT
A strong academic argument goes beyond presenting evidence; it requires analyzing counterarguments and tying observations to broader frameworks. When we tested thesify and ChatGPT on the sample essay, their feedback diverged sharply in depth and applicability.
thesify’s Targeted Feedback:
Deepening Observations:
thesify emphasized the need to clarify the implications of gendered behaviors observed in the test essay. For example, it flagged the organization of bathroom lines by gender and suggested linking this to systemic inequality.
Counterarguments:
thesify identified gaps where counterarguments could strengthen the essay, encouraging a more balanced and nuanced analysis. This included addressing potential critiques of the claims about gendered spaces and their societal implications.
To learn more about the importance of counterarguments, visit Counterarguments in Academic Writing: Why They Matter and How to Approach Them.
Tying Evidence to Broader Frameworks:
By connecting observations, such as costume choices at raves, to sociological concepts like gender performativity, thesify ensured the essay aligned with academic expectations.
Suggesting Related Sources:
One of thesify’s standout features is its ability to suggest academic resources based on highlighted text. For instance, when the phrase "fashion as an identificatory gender display" was highlighted, thesify provided curated resources to explore this concept further. This feature not only saves time but also enriches the depth of analysis by directing writers to relevant, credible materials.
ChatGPT’s Feedback:
Generalized Suggestions:
ChatGPT offered general feedback on the essay’s strengths, such as identifying observations of gender performance, but it failed to address specific gaps in evidence or argumentation
No Resource Integration:
Unlike thesify, ChatGPT did not provide reliable related academic resources or suggestions to enhance the essay with further evidence
No Counterargument Suggestions:
Unlike thesify, ChatGPT overlooked opportunities to suggest counterarguments, leaving the essay less comprehensive and its argument weaker.
Surface-Level Advice:
While it praised the inclusion of certain observations, ChatGPT’s feedback remained generic, lacking the actionable insights needed to refine the essay.
Key Difference:
thesify’s tailored feedback and ability to suggest academic resources allow writers to strengthen their essays with well-supported evidence and deeper connections to theoretical frameworks. ChatGPT, by contrast, lacked the precision and actionable guidance needed to support meaningful improvements.
Summary of Key Differences
| Feature | thesify | ChatGPT |
|---|---|---|
| Ethical Feedback | Encourages critical thinking; avoids text generation. | Generates text, risking academic misconduct. |
| Assignment Alignment | Criteria-based feedback with a clear overall score. | Broad, general feedback; no scoring system. |
| Depth and Specificity | In-depth, actionable feedback tailored to the essay. | Surface-level suggestions; lacks context. |
| Thesis Evaluation | Identifies gaps and provides actionable guidance. | Focuses on style over substance. |
| Academic Resources | Suggests related sources to enhance analysis. | No resource integration. |
Which AI Tool Truly Supports Academic Writing?
Choosing the right AI tool for academic writing means more than just improving your grades—it’s about fostering skills that will serve you throughout your academic and professional journey. While ChatGPT offers convenience, thesify’s feedback emphasizes critical thinking, ethical use, and targeted growth, setting students up for long-term success. If your goal is not just to complete an assignment but to truly excel and learn, thesify provides the structured, purposeful support that students and researchers need in today’s AI-driven world.
Related Blog Posts
9 Tips for Using AI for Academic Writing (Without Cheating)
How AI Is Changing Academic Grading
Writing Excellence in the AI Era: Fostering Academic Writing Skills with Supportive Feedback