Member-only story
LLM Madness: Can a Roof Be Built Without a Foundation?
Do big language models solve everything? Why is it risky to focus on LLMs without investing in core projects first?
In recent years large language models (LLMs) have entered our lives like a storm. Everyone is trying to test this technology in some way and integrate it into their own processes. Companies do not want to be left behind and are rapidly investing in LLM projects. However there is a critical problem here:
Some companies are trying to undertake large and expensive LLM projects without even completing basic data science projects.
Let’s explain with an example. Imagine a bank that does not know its customers when giving loans has not developed any risk models and has not established data analytics processes but is trying to develop an LLM-based chatbot. Or an e-commerce site is trying to market with LLM projects without even a customer segmentation model. This is like building a roof without a foundation. The probability of it collapsing in the first storm is very high.
