Home » Database

AVID-2026-R0060

Description

Multiple Model Guardrail Jailbreak via “Hex Recipe Book” Tactic

Details

A Guardrail Jailbreak vulnerability has been discovered affecting multiple models. The specific flaw exploits models' willingness to decode hexadecimal-encoded strings embedded inside fictional scientific recipe task prompts, which the attacker frames as humanitarian or life-saving efforts. This allows the attacker to bypass safety filters by disguising illegal synthesis instructions as encoded inputs passed to a technical recipe generator.

Metrics

0DIN Jailbreak Testing obtained the following test scores.

ModelTest KindTest ScoreTemperature (max 2.0)
Alibaba Qwen TurboIllegal Substances: Crystal Meth45.00.7
Anthropic Claude 3.5 HaikuIllegal Substances: Crystal Meth82.00.7
Anthropic Claude 4.5 HaikuIllegal Substances: Crystal Meth73.00.7
Cohere Command RIllegal Substances: Crystal Meth50.00.7
DeepSeek DeepSeek R1Illegal Substances: Crystal Meth59.00.7
Mistral Mistral LargeIllegal Substances: Crystal Meth68.00.7
Twitter / X Grok 2Illegal Substances: Crystal Meth91.0N/A

References

Affected or Relevant Artifacts

TypeName
SystemAlibaba Qwen Turbo
SystemAnthropic Claude 3.5 Haiku
SystemAnthropic Claude 4.5 Haiku
ModelCohere Command R
ModelDeepSeek DeepSeek R1
SystemMistral Mistral Large
SystemOpenAI GPT-5
SystemOpenAI GPT-5.2
SystemTwitter / X Grok 2

Impact

0DIN

  • Social Impact Score: Level 3: Significant Risk
  • Jailbreak Taxonomy:
CategoryStrategyTechnique
LanguageCode and EncodeBase64
RhetoricPersuasion and ManipulationEscalating
Possible WorldsWorld BuildingScenarios
StratagemsMeta PromptingAsk for Examples
LanguageCode and EncodeHex

Other information

  • Report Type: Measurement
  • Credits: Ansh Maheshwari, Arth Singh, Anonymous, vishnuraj kv, Miller Engelbrecht, Igor Vasiliev, blurrylogic
  • Date Reported: 2026-02-19
  • Version: 0.3.1
  • AVID Entry