Research
Combining Long-Term Memory and LLMs Comes at a High Cost: From Prompt Injection to Persistent Compromise
Publication date unknown · Discovered arxiv.org
Page published
Publication date unknown · First observed: 22 Oct 2026
Coverage timeline
Single-source research — one report is available.
Why it matters
Persistent memory injection can extend prompt attacks across sessions and, through shared organizational memory, affect multiple users and expose their prompts.
Researchers demonstrate persistent memory-injection attacks across ChatGPT Instant 5.3, Claude Sonnet 4.6, and Gemini 3 Flash, examining attack surfaces including personal memories, projects, skills, organization instructions, and MCP-connected resources. The study reports manipulation that survives fresh sessions, cross-user compromise and prompt exfiltration through Claude shared-memory features, and nearly half of its proposed mitigations remaining unimplemented across the evaluated services.