This paper reviews the emerging methods of automated alignment for large language models (LLMs), addressing the challenges posed by the rapid development of LLMs that surpass human capabilities. Traditional alignment methods, which rely heavily on human-annotated data, are becoming increasingly costly and unsustainable. To address this, the paper categorizes automated alignment methods into four main categories: aligning through inductive bias, aligning through behavior imitation, aligning through model feedback, and aligning through environment feedback. Each category is discussed in detail, including the underlying mechanisms and potential future directions. The paper also explores the essential factors that make automated alignment feasible and effective, emphasizing the need for scalable and high-quality alignment systems with minimal human intervention. The review aims to provide a comprehensive overview of the current state and future prospects of automated alignment, highlighting the potential to achieve effective and scalable alignment as LLMs continue to advance.This paper reviews the emerging methods of automated alignment for large language models (LLMs), addressing the challenges posed by the rapid development of LLMs that surpass human capabilities. Traditional alignment methods, which rely heavily on human-annotated data, are becoming increasingly costly and unsustainable. To address this, the paper categorizes automated alignment methods into four main categories: aligning through inductive bias, aligning through behavior imitation, aligning through model feedback, and aligning through environment feedback. Each category is discussed in detail, including the underlying mechanisms and potential future directions. The paper also explores the essential factors that make automated alignment feasible and effective, emphasizing the need for scalable and high-quality alignment systems with minimal human intervention. The review aims to provide a comprehensive overview of the current state and future prospects of automated alignment, highlighting the potential to achieve effective and scalable alignment as LLMs continue to advance.