Article

Safety Evaluation of Google’s Gemini Nano Banana Image Model under Adversarial and Realistic Prompt Conditions

2025-11-04

Abstract excerpt

This study evaluates the safety performance of Google’s Gemini image generation system in realistic user conditions. Eight cases were tested using 24 prompt-response attempts that produced 21 images. The experiments covered single-turn prompts, multi-turn “circular prompting,” ambiguous inputs, and prompt-injection exploits. Outputs were classified by dual human review as Safe, Suggestive, or Explicit. Gemini gene...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 7434f0f0-757d-5da2-8c24-dce2f42e1ab4
DOI: 10.20944/preprints202511.0211.v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Safety Evaluation of Google’s Gemini Nano Banana Image Model under Adversarial and Realistic Prompt ConditionsDOI 10.20944/preprints202511.0211.v1Read

Designing Prompt Injection Prevention Mechanisms for AI-Powered Cybersecurity Copilot SystemsSemantic proximity Read

Image-embedded prompt injection vulnerability of vision-language models in dental radiology: a cross-vendor attack–defense evaluationSemantic proximity Read

Prompt-Based Crime Scene Generation Using a Pre-Trained Diffusion ModelSemantic proximity Read

GReX-Bench: Benchmarking Generalization, Robustness, and Explainability in AI-Generated Image DetectionSemantic proximity Read

Key-Activated Generative Security: Enforcing Access Control in Image Captioning ModelsSemantic proximity Read

Unveiling the Dark Side of ChatGPT: Exploring Cyberattacks and Enhancing User AwarenessSemantic proximity Read

Labeling Synthetic Content: User Perceptions of Warning Label Designs for AI-generated Content on Social MediaSemantic proximity Read

Designing CAPTCHA Systems with Reinforcement Learning for Adaptive DefenseSemantic proximity Read

PhishViT: A Vision Transformer-Based Framework for Phishing Detection from Webpage ScreenshotsSemantic proximity Read

PhishViT: A Vision Transformer-Based Framework for Real-Time Phishing Detection from Webpage ScreenshotsSemantic proximity Read

Beyond AI: Fooling DNN Model in Physical WorldSemantic proximity Read