Trojan Horse Prompting: Jailbreaking Conversational Multimodal Models by Forging Assistant Message
This exposes a fundamental flaw in conversational AI security, necessitating a paradigm shift for developers and researchers to enhance safety protocols.
The paper tackled the vulnerability of conversational AI models to jailbreak attacks by introducing Trojan Horse Prompting, which forges assistant messages in dialogue history to bypass safety mechanisms, achieving a significantly higher Attack Success Rate than existing methods on Google's Gemini-2.0-flash-preview-image-generation.
The rise of conversational interfaces has greatly enhanced LLM usability by leveraging dialogue history for sophisticated reasoning. However, this reliance introduces an unexplored attack surface. This paper introduces Trojan Horse Prompting, a novel jailbreak technique. Adversaries bypass safety mechanisms by forging the model's own past utterances within the conversational history provided to its API. A malicious payload is injected into a model-attributed message, followed by a benign user prompt to trigger harmful content generation. This vulnerability stems from Asymmetric Safety Alignment: models are extensively trained to refuse harmful user requests but lack comparable skepticism towards their own purported conversational history. This implicit trust in its "past" creates a high-impact vulnerability. Experimental validation on Google's Gemini-2.0-flash-preview-image-generation shows Trojan Horse Prompting achieves a significantly higher Attack Success Rate (ASR) than established user-turn jailbreaking methods. These findings reveal a fundamental flaw in modern conversational AI security, necessitating a paradigm shift from input-level filtering to robust, protocol-level validation of conversational context integrity.