PSA: The reason why it feels like ChatGPT is less smart nowadays is because of the following three reasons.

Courtesy of a redditor who works at Open AI<br><br>1) **The training data is 4 months old**: The training data for the base model hasn’t been updated since December of last year. Which, for a machine learning model, is a very long time. So any information that has been found or announced since December (like any of the irregularities that have been found in the latest COVID vaccines or any developments in cancer research) it won’t know about<br><br>2) **It doesn’t have access to the internet**: It’ll only know what it has been trained on. It is not capable of looking up anything on the internet. So if it doesn’t know something it cannot look it up. If it’s wrong about something it cannot look up the correct answer. The only thing it has is what it has been given. <br><br>3) **It doesn’t have any new updates**: None of the updates from inside Open AI have been rolled out to ChatGPT since December. So all the improvements they have made to the model in the past few months, it doesn’t have access to. Which means it is worse than the current version of the model.