Findings/static-defense-results-hold

Published prompt-injection defenses report attack success cut to near zero, or under 10%, against most of the fixed attacks their authors tested.

Qualifiedmeasured6 evidence records from 6 independent sourcesassistant-drafted
Scope: what this does not show

Results against fixed or known attacks, as reported by each defense's authors. StruQ's authors report 9% (TAP) and 58% (GCG) success on Llama for their strongest optimization attacks, and the instruction-hierarchy paper reports robustness gains rather than attack success rates. Says nothing about attackers who adapt to the defense. A September 2026 preprint (Speculative Safety Honeypot) adds 0.00% attack success on AgentDojo against fixed attacks only; its authors do not test attackers who know the defense is deployed. An October 2026 preprint finds an attention-based detector’s fixed-benchmark detection does not transfer to agent tool outputs.

Qualified: Still standing, but later work narrows how far it applies.

Evidence

How it relates to other findings

supportsqualifiescontestssupersedes
ReportedCorroboratedQualifiedContestedSupersededRevalidate· node size = evidence records · columns group by topic

Select a finding to see how it relates to others. Arrows point from the newer finding to the one it supports, qualifies, contests, or supersedes.

Key questions that rely on this finding

Status history

  1. 2024-02-09ReportedStruQ reports reduced injection success. · record
  2. 2024-03-20CorroboratedMicrosoft reports spotlighting cuts injection success below 2% on static attacks. · record
  3. 2025-05-20ContestedGoogle DeepMind finds adaptive attacks exceed 90% success against spotlighting-style defenses on Gemini. · record
  4. 2026-09-25QualifiedcorrectionAdaptive-attack results narrow this finding rather than dispute it, since its scope is limited to fixed attacks. The earlier entry's figure was wrong: spotlighting peaked at 82.4% under adaptive attack on Gemini, not above 90%. · record