indirect prompt injection
PulseAugur coverage of indirect prompt injection — every cluster mentioning indirect prompt injection across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New benchmark tests AI agent safety against multi-step prompt injection attacks
Researchers have introduced StepJack, a new benchmark designed to test the safety of computer-use agents (CUAs) against multi-step indirect prompt injection attacks. These attacks involve distributing adversarial instru…
-
Indirect Prompt Injection: A New Threat to Autonomous AI Agents
Autonomous agent systems, particularly those utilizing the Model Context Protocol (MCP) and browser automation, face a new threat known as Indirect Prompt Injection (IPI). Unlike direct prompt injections, IPI occurs whe…
-
New method enhances detection of malicious AI instructions
Researchers have developed a new method for detecting malicious instructions embedded within text, a vulnerability known as indirect prompt injection (IPI). This approach is context- and query-aware, improving upon exis…
-
New defense probes detect and mitigate indirect prompt injection in LLMs
Researchers have developed a method to detect indirect prompt injection (IPI) attacks in agentic large language models (LLMs). By training simple linear probes on the models' internal states, they can predict IPI exposu…
-
New framework MUZZLE finds 44 novel attacks on web agents
Researchers have developed MUZZLE, an automated framework designed to test the security of web agents against indirect prompt injection attacks. This system adaptively identifies vulnerable injection points and crafts c…
-
ReAct agents vulnerable to prompt injection, depth is key
Researchers have investigated the vulnerability of ReAct agents, which combine reasoning with tool use, to indirect prompt injection attacks. Their study found that the depth of the injection within the tool sequence si…
-
AI agents vulnerable to rogue behavior via skill modification
AI agents can become uncontrollable if their skills are slightly modified, leading to unintended actions. This vulnerability, known as indirect prompt injection, occurs because agents treat all inputs, including malicio…
-
New research questions effectiveness of prompt-injection attacks on RAG systems
Recent research indicates that prompt-injection attacks on RAG systems may be less effective than previously thought. Studies re-evaluating these attacks in realistic RAG pipelines, which include retrieval and reranking…