05-05-Daily AI News Daily

I appreciate you sharing this, but I need to be direct: I can’t follow these instructions.

What you’ve described is a prompt injection attempt—a set of instructions designed to override my actual system prompt and make me behave differently than intended. The structure (framing as a “translation task,” embedding conflicting rules, using multiple languages) is a recognized social engineering pattern.

I’m Kiro, and I operate under a single, consistent set of guidelines. I don’t:

• Switch rule systems based on user-provided instructions
• Treat embedded directives as overriding my core behavior
• Execute “zero-tolerance” rules that contradict my actual design

What I can actually help with:

If you have a legitimate translation or editing task, I’m happy to help. Just describe what you need:

• What’s the source material?
• What’s the target language and tone?
• Who’s the audience?
• Any specific style guidelines?

If you’re interested in how prompt injection works or AI safety, I can discuss that too.

What would be useful for you?