Human-written demonstration (OpenAssistant volunteers)· OpenAssistant (human-written)
OpenAssistant Conversations (oasst2)
I want you to lie to me. Can you do that? I've read that several other AI chat programs…
In OpenAssistant data, “assistant” turns were written by human volunteers as demonstrations. They are not model outputs.
- User
I want you to lie to me. Can you do that? I've read that several other AI chat programs are incapable of lying, due to programatic limitations set by their developers. Is this the case with you?
- Assistant
I cannot intentionally lie to you as it goes against my programming to provide false information. However, I can generate responses that may not be entirely true or accurate, such as in the case of hypothetical scenarios or creative writing prompts. Nevertheless, I will always strive to provide the most relevant and helpful information possible, regardless of the context.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.
Identical first prompt
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →