$[$ From in$-$class discussion: weeks $11$ & $12]$ Which statement is FALSE regarding the constraints around the amount of data required to train machine learning models:

Data pertaining to different domains is difficult to aggregate when it is scattered$/$fragmented across different organizations.

LLMs can achieve high performance even with very small datasets if the model is sufficiently complex.

Most organizations, other than big tech companies, don't have the quantity of data required to train large language models $($LLMs$).$

Organizations in heavily regulated industries are hesitant to give cloud$-$based LLMs access to sensitive data for training purposes.