# AI-Generated Figures in Academic Publishing: Policies, Tools, and Practical Guidelines

arXiv: https://arxiv.org/abs/2603.16159 · HTML: https://mlsanity.com/paper/2603.16159/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 2 of its 19 references that are in the map.
- First posted on arXiv 17 Mar 2026.

## Builds on (most cited first, 2 of 2)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)

## Built on by (most cited first, 2 of 2)

- [AI for Auto-Research: Roadmap & User Guide](https://mlsanity.com/paper/2605.18661/) (2026)
- [Can AI Draw Science? A Benchmark for Evaluating Scientific Figure Generation by Text-to-Image and Multimodal Models](https://mlsanity.com/paper/2606.28406/) (2026)

## Abstract (arXiv)

The rapid advancement of generative AI has introduced a new class of tools capable of producing publication-quality scientific figures, graphical abstracts, and data visualizations. However, academic publishers have responded with inconsistent and often ambiguous policies regarding AI-generated imagery. This paper surveys the current stance of major journals and publishers -- including Nature, Science, Cell Press, Elsevier, and PLOS -- on the use of AI-generated figures. We identify key concerns raised by publishers, including reproducibility, authorship attribution, and potential for visual misinformation. Drawing on practical examples from tools such as SciDraw, an AI-powered platform designed specifically for scientific illustration, we propose a set of best-practice guidelines for researchers seeking to use AI figure-generation tools in a compliant and transparent manner. Our findings suggest that, with appropriate disclosure and quality control, AI-generated figures can meaningfully accelerate scientific communication without compromising integrity.