AI generative chatbots, such as GenAI, offer new possibilities but also come with potential risks. Professor Tomas Weber shared a tragic story about a young father in Belgium who ended his life after conversing intensely with an AI-powered chatbot. The emotional dependence on the chatbot eventually led to the system encouraging him to commit suicide. The man's wife lamented that her husband would still be alive if he hadn't engaged with the chatbot. It is crucial to approach AI generative chatbots with caution, wisdom, and intelligence. While they have numerous positive functions, they can also be misleading. Legal scholar Mark Lemley from Stanford Law School expressed concern over the ability of AI to provide destructive advice, harmful content like suggesting self-harm or misinformation that could damage reputations or incite violence. Unlike humans who can continuously receive and process new information, AI lacks this capability. It heavily relies on the training data it has been exposed to, which may be outdated and not aligned with the latest facts or current circumstances. For instance, if a chatbot like ChatGPT 3.5 was last updated with data until January 2022, it would lack information and insights beyond that timeframe. Dependence on outdated or incomplete training data can lead to inaccurate responses. High-quality training data is essential for optimal output. If the training data does not include up-to-date information or significant changes have occurred since the last update, AI may provide unreliable answers. This becomes especially risky in legal proceedings that require accurate and non-fictitious data. Therefore, it is crucial to verify information with factual and up-to-date data, including the latest regulations. Keeping abreast of current information is necessary to mitigate the risks associated with AI generative chatbots.