# Prompt-in-Content Attacks: Exploiting Uploaded Inputs to Hijack LLM Behavior

Source: [Prompt-in-Content Attacks: Exploiting Uploaded Inputs to Hijack LLM Behavior](<https://www.sota2.com/research/paper/68fa7add1e209131cea85970>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T05:28:46.355Z

Authors: Zhuotao Lian, Weiyu Wang, Qingkui Zeng, Toru Nakanishi, Teruaki Kitasuka, Chunhua Su

Published: 2025-08-25T05:20:11.000Z

## Abstract

Large Language Models (LLMs) are widely deployed in applications that accept user-submitted content, such as uploaded documents or pasted text, for tasks like summarization and question answering. In this paper, we identify a new class of attacks, prompt in content injection, where adversarial instructions are embedded in seemingly benign inputs. When processed by the LLM, these hidden prompts can manipulate outputs without user awareness or system compromise, leading to biased summaries, fabricated claims, or misleading suggestions. We demonstrate the feasibility of such attacks across popular platforms, analyze their root causes including prompt concatenation and insufficient input isolation, and discuss mitigation strategies. Our findings reveal a subtle yet practical threat in real-world LLM workflows.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2508.19287>)

[PDF](<https://arxiv.org/pdf/2508.19287.pdf>)
