# A New Paper Argues Prompt Injection Cannot Be Fully Defended in Today's Architectures

Researchers offer a proof that models which share one embedding space for instructions and data lack the separation any robust defense would require.

- Published: 2026-06-29T10:45:14.906Z
- Canonical: https://polylog.news/ai/2026-06-29/a-new-paper-argues-prompt-injection-cannot-be-fully-defended
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [arXiv (cs.CR)](https://arxiv.org/abs/2606.27567), [Anthropic Frontier Red Team](https://www.anthropic.com/research/team/frontier-red-team)

A paper posted to arXiv, On the Inseparability of Instructions and Data in Shared-Embedding Sequence Models, takes up what its authors call the top security risk for applications built on large language models. Every prompt-injection defens…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-06-29/a-new-paper-argues-prompt-injection-cannot-be-fully-defended (subscription information: https://polylog.news/pricing).