https://hackerpaper.kaipfister.de/p/2504.21228/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=44507977urn:hackerpapers:mention:2504.21228:44507977Wed, 09 Jul 2025 09:43:35 GMT<p>Academic researcher here working on this exact issue. Prompt engineering methods are no sufficient to address the challenge. People in Academy and Industry labs are aware of the issue and actively working on it, see for instance: [1] Camel: work by google deepmind on how to (provably!) prevent agent planner from being prompt-injected: https://github.com/google-research/camel-prompt-injection [2] FIDES: similar idea by Microsoft, formal guarantees: https://github.com/microsoft/fides [3] ASIDE: marking non-executable parts of input and rotating their embedding by 90 degrees to defend against…</p><p><a href="https://hackerpaper.kaipfister.de/p/2504.21228/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CR