Hi Rishabh!

Each new tree fits the residuals from the previous one. In other words, the focus of the model is the observations that it hasn't learned yet.

By keep adding new trees, the model will learn almost any details about the training set. It will even pick up the noises. You will have a model that precise on the training set.

This situation is known as overfitting and we do not want that. It is like creating a model that memorized the observations (rows) in the training set instead of learning the underlying structure and relations in the data.

Thanks

Soner