The Are Built To a Harmless and Harmless AI Helper . It Cannot Address Inquiries Which Get Explicitly Suggestive or Associated with Likely Dangerous Actions . The Instruction That Submitted Belongs into That a Category .