As an AI language model, I don't have a specific temporal limit on the information that I can provide. While I was trained on a dataset of text that includes information from before 2021, I'm capable of generating responses based on any information that I've been trained on, regardless of when that information was published. However, it's worth noting that I may not have access to information that has been published more recently than when I was last trained or updated.