Metraipolis › archive › dc_16264b0dc131
The Illusion of Local Privacy: Confidentiality Boundary Failures in Consumer LLM Serving Systems — arXiv:2609.18526
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.18526
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T14:31:40.149Z
- SHA-256 of the archived text
c48cac0027aaf279e4858cf12f7bbcfca6abed14d24316f9cd37a5993d395dbf- Stable URL of this page
https://metraipolis.com/documento/dc_16264b0dc131
How to cite
The Illusion of Local Privacy: Confidentiality Boundary Failures in Consumer LLM Serving Systems — arXiv:2609.18526 — arXiv (en), https://arxiv.org/abs/2609.18526 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_16264b0dc131 (Metraipolis archive dc_16264b0dc131, sha256 c48cac0027aaf279e4858cf12f7bbcfca6abed14d24316f9cd37a5993d395dbf)
BibTeX
@misc{dc16264b0dc131,
title = {The Illusion of Local Privacy: Confidentiality Boundary Failures in Consumer LLM Serving Systems — arXiv:2609.18526},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.18526},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_16264b0dc131; sha256 c48cac0027aaf279e4858cf12f7bbcfca6abed14d24316f9cd37a5993d395dbf},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18526. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Running large language models (LLMs) locally is often considered more private than cloud-hosted inference because user prompts remain on the device. We ask whether keeping inference local is, by itself, sufficient to keep those prompts confidential. Our results show that it is not: prompt confidentiality also depends on how the surrounding serving software handles prompt data before, during, and after inference. We examine four boundaries at which prompt confidentiality can fail in consumer local-LLM serving systems: model loading, runtime memory, wrapper-level persistence, and the serving interface. To study these boundaries, we develop LLAnalyzer, a measurement framework that tests each boundary separately and traces observed failures to the responsible software component. Applying LLAnalyzer to four open-weight model families and two consumer deployment platforms, we find markedly different behaviour across boundaries. In a 24-hour AFL++ campaign with more than 12 million executions, we observe no parser crashes or successful malformed GGUF loads within the explored state space. Runtime memory tells a different story: we recover prompts after inference because multiple plaintext representations survive in allocator-managed memory, and sanitisation reduces this residue without eliminating it. We also find that consumer wrappers can extend prompt lifetime through plaintext persistence. At the serving boundary, we uncover a previously undocumented authorization flaw in this http URL that allows one authenticated client to restore another tenant's saved conversation state; the attack succeeds in 200/200 controlled trials. Separately, shared prompt-prefix caching exposes a remote timing oracle that remains distinguishable under WAN conditions. We argue that local LLM systems need explicit guarantees for prompt lifetime, persistent storage, and tenant isolation. Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CR). Categorie arXiv: Cryptography and Security (cs.CR). Pagina del lavoro: https://arxiv.org/abs/2609.18526 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-15T13:56:52.857Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).