This systematic review explores the current applications of large language models (LLMs), particularly ChatGPT and GPT-4, in psychiatry. The review follows PRISMA guidelines and searches multiple databases up to March 2024. From 771 retrieved articles, 16 studies were included that directly examined LLMs' use in psychiatry. These studies highlight diverse applications of LLMs, including clinical reasoning, social media, and education. LLMs can assist in diagnosing mental health issues, managing depression, evaluating suicide risk, and supporting educational interventions. However, limitations include difficulties with complex cases and potential underestimation of suicide risks. The review concludes that while LLMs show promise, they are not yet ready for broader clinical use, and future research should focus on their integration into mental health care, balancing benefits and risks.This systematic review explores the current applications of large language models (LLMs), particularly ChatGPT and GPT-4, in psychiatry. The review follows PRISMA guidelines and searches multiple databases up to March 2024. From 771 retrieved articles, 16 studies were included that directly examined LLMs' use in psychiatry. These studies highlight diverse applications of LLMs, including clinical reasoning, social media, and education. LLMs can assist in diagnosing mental health issues, managing depression, evaluating suicide risk, and supporting educational interventions. However, limitations include difficulties with complex cases and potential underestimation of suicide risks. The review concludes that while LLMs show promise, they are not yet ready for broader clinical use, and future research should focus on their integration into mental health care, balancing benefits and risks.