Sep 10, 2025
The rise of advanced editing software and AI technolodgy has made safeguarding image integrity more critical than ever in scientific publishing. Now adopted by MDPI in selected journals, Proofig AI strengthens peer review by combining automated analysis with human oversight. With advanced image proofing features, the platform saves time, reduces human error, and supports researchers, editors, and publishers in upholding the highest standards of integrity.
With the rapid advancement of digital editing and AI technologies, the integrity of scientific images has become a growing concern across the research community. Modern image editing tools and generative AI models can produce convincing but misleading visuals, making it crucial for publishers and researchers to adopt advanced detection methods. Addressing this need, Proofig AI provides an automated, AI-powered solution designed to safeguard image authenticity in scientific publications — a system now being implemented across selected MDPI journals to reinforce editorial quality and transparency.
The Rising Challenge of Image Integrity
Manipulated or duplicated images pose a significant threat to scientific credibility. According to the International Association of Scientific, Technical and Medical Publishers (STM):
“Image alteration and/or duplication can be much more damaging, as it corrupts actual research results, wastes research money on invalid leads, undermines society’s trust in research, and can even endanger the society in which those ‘results’ are used.”
Digital editing tools and AI generators have made such manipulations easier to produce — whether deliberate or accidental. As Tim Tait-Jamieson, Head of Publication Ethics at MDPI, explains:
“Images are a core part of articles, conveying important information, concepts, and strengthening the conclusions.”
When an image is altered or misrepresented, it can distort the data, mislead readers, and compromise the reliability of findings. Even unintentional mistakes can have lasting consequences, which is why maintaining image integrity is fundamental to maintaining research credibility.
Common Types of Image Integrity Issues
Proofig AI categorizes image integrity risks into several key areas:
Image duplication: Reusing identical or partially identical images in different contexts.
Manipulation: Altering data or visuals to misrepresent results.
Fabrication: Creating entirely false or synthetic images.
Plagiarism: Using another researcher’s images without permission or citation.
Improper handling: Including poor labeling, low-quality visuals, or misused image files.
Misrepresentation: Drawing incorrect conclusions or interpretations from visuals.
Each of these can distort the interpretation of data and undermine the reliability of scientific findings.
Responsible Use of AI in Research
Regarding AI-generated visuals, Tim Tait-Jamieson notes:
“Sometimes authors will design a concept figure for their article to clarify a process or specific methodology. Such concept figures require artistic elements that do not represent scientific data and may be created using AI tools. However, at MDPI, we discourage such use of AI tools as they may produce scientifically inaccurate or even plagiarized figures.”
He adds:
“Authors are always responsible for verifying the scientific accuracy of any content generated by generative AI tools and should also ensure no copyrighted material appears in the final image.”
At MDPI, the use of generative AI is strictly prohibited for creating or enhancing research results, including images, photographs, measurements, and visualizations such as graphs or blots.
Avoiding Image Manipulation
Researchers are encouraged to maintain replicable workflows, adhere to publisher guidelines on AI use, and preserve raw image data and equipment details. Editors and reviewers also play a vital role in verification. As Tim Tait-Jamieson outlines:
“Reviewers should focus on the clarity, accuracy, and relevance of the images to the research question and conclusions. This includes assessing whether the images effectively support the presented findings, if the methodology used to generate them is appropriate, and if any ethical concerns exist.”
He emphasizes that reviewers should check labeling, consistency, and access to raw data where available. Manipulated images, he notes, might stem from weak underlying data or misrepresentation, and such issues must be addressed early in the review process.
How Proofig AI Strengthens Image Verification
Proofig AI uses advanced AI and pattern-recognition algorithms to detect plagiarism, duplication, and manipulation within images. Its AI Image Fabrication feature is designed to identify unique digital patterns commonly found in AI-generated visuals, flagging synthetic images for review.
According to Gabriel Martinez, Scientific Advisor at MDPI:
“From my experience, what makes Proofig AI unique compared to other tools is its user-friendly interface and its ability to efficiently detect image duplications and manipulations with high accuracy. Furthermore, Proofig AI offers visual reports that help identify problematic images.”
The system compares figures both within the manuscript and against published literature indexed in PubMed, providing similarity scores and image-matching insights. These visual reports save time and reduce human error, allowing editors to review flagged content with confidence.
Integrating Proofig AI into MDPI’s Editorial Workflow
When a paper is submitted, MDPI’s editorial team assesses whether the included image types are suitable for Proofig AI screening. The manuscript is then uploaded either directly into Proofig’s web interface or through an integrated API. Each detected match is evaluated by staff members using similarity scores and filters that highlight potential manipulations such as resizing or rotation.
Oliver Terrett, Scientific Advisor, explains:
“A staff member must then manually review every match in Proofig AI. Each match provides a similarity score, and other relevant info (such as whether the image has been rotated or re-sized). … Comments can be added for each match, and each verified match can be assembled into a PDF report which can be shared with other staff or Editorial Board Members.”
Proofig AI’s continuous learning system refines its accuracy over time, adapting to new manipulation methods and image-generation technologies.
Building a Future of Transparent Peer Review
MDPI’s Head of Technology Innovation, Milos Cuculovic, highlights how this initiative aligns with the publisher’s broader technological goals:
“In 2018, we established the Technology Innovation team, composed of AI experts including data scientists, AI/ML engineers, and data engineers. … From the experts, it seemed Proofig AI was by far the best choice, and here we are: further ensuring research integrity and guaranteeing the high quality of research output published by MDPI.”
To ensure consistent usage, MDPI has issued detailed Proofig AI user guidelines for editors and reviewers, outlining parameters for optimal detection and guidance for investigating potential issues according to COPE (Committee on Publication Ethics) standards.
As Gabriel Martinez adds, these protocols help users “make good use of this type of tool,” ensuring that AI assists — but never replaces — expert human judgment.
Human Expertise Enhanced by AI
While automation significantly improves efficiency, human oversight remains central to MDPI’s editorial standards. The integration of Proofig AI exemplifies how AI tools can enhance — not replace — expert review.
By combining machine learning with editorial scrutiny, MDPI and Proofig AI aim to ensure that every published figure reflects genuine, trustworthy research. This partnership strengthens the foundation of research integrity, reinforcing confidence in scientific publishing in an era of increasing digital and AI-driven challenges.