While citations generated by ChatGPT may look convincing, they are results of aggregated textual information rather than evaluated information.
Response by OpenAI to a query to better understand the parameters within which ChatGPT-3 operates:
As an AI language model, I do not have the ability to access or browse the internet like humans do. My responses are based on the text input I receive from users, and my knowledge is derived from the vast amount of text data that I have been trained on.
The training data used to create my base architecture, GPT-3, included a diverse range of text sources, such as books, articles, websites, and other online content, covering a time span from the 1980s to 2020; with a knowledge cutoff of September 2021.
However, it's worth noting that my responses are based on the information that was available in the text corpus at the time of my training, and my knowledge is not constantly updated in real-time. As new information and events occur, my responses may not always reflect the latest developments, and it's possible that I may not have knowledge of recent events or changes.
Reference
OpenAI. (2023). ChatGPT-3 (April 12, 2023 version) [Large language model]. https://chat.openai.com/chat