Five core principles now guide OpenAI as it pursues artificial general intelligence. Sam Altman emphasizes broad access while explicitly reserving the right to restrict specific capabilities. This framework attempts to balance open utility with risk mitigation. Practitioners should monitor how these guidelines translate into actual deployment constraints for future frontier models.