In this post, we examine the performance of Manus’s newly updated slide generation tool when applied to a peer-reviewed scientific article. The developers claim recent improvements focused on enhancing the tool’s ability to support academic communication. To test these capabilities, we selected a published study in political science and prompted the tool to create a 20-slide presentation suitable for a 30-minute university lecture. We aimed to explore whether Manus can transform a complex academic text into a clear, structured, and content-rich presentation—without oversimplifying the arguments or losing the study’s conceptual depth.
Input file
As input, we used our own peer-reviewed article titled The Concept of Tailor-Made Laws and Legislative Backsliding in Central–Eastern Europe (Kiss, Rebeka and Sebők, Miklós, Comparative European Politics, 2025).
Prompt
To evaluate the slide generation capabilities of Manus (manus.ai), we provided a structured prompt instructing the model to generate a university-level PowerPoint presentation based on the uploaded academic article. The aim was to produce a 20-slide deck suitable for a 30-minute scholarly lecture, using clear and concise UK English. The requested structure followed standard academic presentation conventions—from the research motivation and theoretical framework to data & methods, findings, and conclusions—and required content to be accessible to an interdisciplinary audience while retaining analytical rigour. Visual elements and bullet-point formatting were encouraged to enhance clarity and engagement.
You are a professional academic presentation assistant. Please generate a PowerPoint presentation (.pptx format, 16:9 aspect ratio) based on the attached academic study. The goal is to create a 30-minute academic talk, summarising and presenting the study in a clear, structured, and engaging way for a university-level audience.
The presentation should include 20 slides, and must be written in UK English. Each slide should have a clear and concise title and well-structured bullet points (avoid full paragraphs). If helpful, include simple charts or visuals based on the study’s data.
The presentation must follow this structure:
- Title Slide (title, author(s), affiliation)
- Motivation and Background
- Research Problem
- Research Questions and Hypotheses
- Theoretical Framework
- Literature Review Highlights
- Research Design
- Methodology Overview
- Data Sources
- Data Collection and Sampling
- Variables and Operationalisation
- Analytical Strategy
- Key Findings – Overview
- Key Findings – Detail I
- Key Findings – Detail II
- Discussion
- Implications
- Limitations
- Conclusion
- Thank You / Q&A slide
Ensure the content is accessible to an interdisciplinary academic audience. Avoid jargon unless explained, and highlight the contribution and novelty of the study.
Output
Manus produced a presentation that broadly reflects the expected academic structure, with slides covering theoretical, methodological, and empirical aspects of the study. However, the slide content at times lacked clarity and depth. Some of the key concepts were oversimplified or loosely reformulated, and important distinctions made in the original text were not always preserved.
The slides are not formatted in 16:9 ratio, and the content frequently overflows beyond the visible area. Key information spills off the slide, compromising both readability and visual coherence.
Recommendations
While Manus’s slide generator reproduces a standard academic structure and applies a clean visual layout, the output lacks conceptual depth and coherence. Key terms are often simplified, slide content frequently exceeds visible space, and the presentation fails to preserve the analytical clarity of the original article. For now, we do not recommend using Manus for generating scientific presentations that aim to meet academic standards.
The authors used Manus [General-purpose AI agent (accessed on 17 June 2025), available at: https://manus.ai] to generate the output.
