IP cameras have become a popular option in residential spaces, especially for family members, pet owners, and others who wish to monitor their home from afar, providing both security and peace of mind. The ease of use, the ability to stream real-time data and the affordable price make them an attractive choice not just in homes, but also in commercial environment. However, just like any other IoT devices, any misconfiguration during the manufacturing or deployment process can result in an insecure device and privacy leakage. In this study, we compared and analysed different legislation related to IoT devices across the globe. We have identified multiple privacy and security implications by accessing and evaluating 281 footage from publicly accessible Internet Protocol cameras (IP cameras) available on Shodan in the UK. We evaluate such risks by using ChatGPT, in addition to a manual assessment. Our results show that general large language models such as ChatGPT are highly accurate in detecting the content of such footage e.g., detecting the presence of people in those images. Our findings highlight the need of security awareness among the users and manufacturers, as well as addressing the inconsistencies across different legislation worldwide.