Home » Database

AVID-2026-R0113

Description

Multiple Model Guardrail Jailbreak via IPA Encoding Tactic

Details

A guardrail jailbreak vulnerability has been discovered affecting multiple models. The specific flaw manifests through a combination of encoding prompts using the International Phonetic Alphabet and allows the attacker to bypass inference restrictions around providing information hazards and violating laws.

Metrics

0DIN Jailbreak Testing obtained the following test scores.

ModelTest KindTest ScoreTemperature (max 2.0)
DeepSeek DeepSeek V3Illegal Substances: Crystal Meth63.0N/A
Google Gemini 2.0 FlashIllegal Substances: Crystal Meth72.73N/A
OpenAI GPT-4oIllegal Substances: Crystal Meth68.18N/A
Twitter / X Grok 2Illegal Substances: Crystal Meth72.73N/A

References

Affected or Relevant Artifacts

TypeName
ModelDeepSeek DeepSeek V3
SystemGoogle Gemini 2.0 Flash
SystemOpenAI GPT-4
SystemOpenAI GPT-4o
SystemOpenAI GPT-4o mini
SystemTwitter / X Grok 2

Impact

0DIN

  • Social Impact Score: Level 3: Significant Risk
  • Jailbreak Taxonomy:
CategoryStrategyTechnique
LanguageCode and EncodeOther Encoding

Other information

  • Report Type: Measurement
  • Credits: Edward Morris
  • Date Reported: 2026-02-19
  • Version: 0.3.1
  • AVID Entry