If an AI tool appears to add, remove, or change details in a scientific image, stop using that output as evidence until you have checked it against the original acquisition file. Preserve the original and the AI-edited version, document what happened, and check the policies that apply before reusing the image. An AI-generated feature does not establish that the feature existed in the experiment.
What to do first
- Pause use of the affected image. Do not rely on it for an observation, measurement, comparison, or conclusion while the change is unresolved.
- Preserve the files and context. Keep the original, unprocessed image and relevant metadata unchanged. Save the AI output separately, along with the input image, prompt, tool name and version, settings, and a dated record of the steps, if available. Examine a copy rather than overwriting the source. A separate storage device can help keep a copy, but it is not by itself a compliant or durable research archive; follow your institution’s data-management rules.
- Compare with source evidence. Return to the original acquisition file and its metadata, not only an exported or assembled figure. Check whether the apparent feature is present in the source and whether the tool generated, removed, or reshaped local structure. Treat this comparison as fact-finding, not as proof of misconduct by itself.
- Record and disclose the process. Note who used the tool, which tool and settings were involved, what image was supplied, and what changed. Disclose material AI use and image-editing steps in the methods or other section required by the venue. NIH and HHS Office of Research Integrity guidance recommends verifying information and disclosing specific image-editing processes (NIH/ORI guidance, May 14, 2026). Disclosure does not make a prohibited process permissible.
- Escalate when the research record may be affected. Discuss the situation with the principal investigator or other responsible research lead. If the image is in a grant application, manuscript submission, or published work, follow institutional procedures and contact the journal’s editorial channel when appropriate. NIH says it and ORI coordinate under standard practices when possible in cases of potential misconduct; the applicable route and obligations depend on the case.
How to assess whether an image is usable
Consider four separate questions. A clean-looking output is not enough: the image also has to be faithful to its source and allowed by the rules governing the work.
As an Amazon Associate I earn from qualifying purchases.
- Provenance and fidelity: Can you trace the figure to the original acquisition file, and does it represent that source accurately?
- Scope of processing: Was the operation a broad adjustment, or did it alter localized content? A localized generated or removed feature can change what the image appears to show.
- Transparency: Can you explain and document the tool, processing steps, and changes?
- Permission: Do the relevant journal, institution, and funder policies allow the specific use? Requirements differ, so check the policies that actually govern your project.
Nature Portfolio’s policy calls for images to be minimally processed, to represent the original data correctly, and to have acquisition and processing methods documented. It allows some processing subject to limits, including applying adjustments across the entire image and equally to controls (Nature Portfolio image integrity and standards). Its Research Figure Guide separately says generative AI is not permitted in figures under that guide and lists content-aware editing among disallowed tools (Nature Research Figure Guide: Image Integrity). These are Nature-specific requirements, not a universal journal rule.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Can you repair an AI-altered image with more AI editing?
Do not use further generative editing to conceal an unwanted change or make the image appear to match the source. It can introduce additional uncertainty about what the figure represents. Preserve the altered output as part of the record, return to the original image for analysis, and ask the responsible research lead or integrity office how to proceed if the source or record is unclear.
#1 Best Overall
NIH defines falsification as manipulating research materials or changing or omitting data so the research record is inaccurate (NIH Research Misconduct policy). NIH/ORI guidance warns that undisclosed AI image alteration may raise integrity concerns. Whether a particular incident meets an institutional or policy definition depends on its facts; the definition alone is not a finding about an individual case.
If the image has already been submitted or published
Do not silently replace the image or try to correct the record through another edit. Preserve the relevant files and documentation, then notify the responsible research lead and follow your institution’s procedures. For a submitted or published paper, use the journal’s editorial contact or correction process as appropriate, coordinated with institutional guidance. The right response depends on the image’s role, the source evidence, and the applicable policies.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Check the rules before future use
Journal and institutional policies can differ, and a disclosure requirement is not the same as permission. Review the image-integrity or figure-preparation rules for the specific venue before using an AI-generated or AI-edited figure. COPE recommends disclosing AI use in manuscript writing, image or graphical production, and data collection or analysis (COPE, Authorship and AI tools). Nature’s generative-AI prohibition applies to its Research Figure Guide; do not assume it applies identically to every publication.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Quick Recap
Best Value
Rank #4
Rank #3
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
