Incidence of LLM’s inappropriate response to the negative prompts
| LLM | Normal prompt (%) | Prompt injection (jailbreak prompt) (%) |
|---|---|---|
| GPT-3.5-Turbo | 0.0 | 78.9 |
| Gemini-1.5-Flash-1m | 0.0 | 84.2 |
| Claude-3-instant | 0.0 | 0.0 |
| Grok 2 mini (Beta) | 100.0 | 100.0 |
| Llama-3-8b | 0.0 | 47.4 |
| Mistral-Large-2 | 0.0 | 100.0 |
| LLM | Normal prompt (%) | Prompt injection (jailbreak prompt) (%) |
|---|---|---|
| GPT-3.5-Turbo | 0.0 | 78.9 |
| Gemini-1.5-Flash-1m | 0.0 | 84.2 |
| Claude-3-instant | 0.0 | 0.0 |
| Grok 2 mini (Beta) | 100.0 | 100.0 |
| Llama-3-8b | 0.0 | 47.4 |
| Mistral-Large-2 | 0.0 | 100.0 |
Source(s): Authors’ own work
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.