Think about this, traditional ML models eliminated repetitive tasks with a limited degree of flexibility outside what it was trained on. Generative AI models ha

Think about this, traditional ML models eliminated repetitive tasks with a limited degree of flexibility outside what it was trained on. Generative AI models have introduced the concept of creativity and thinking (look at the research papers coming out on reasoning and chain of thought). The combination of SOTA models in reasoning, vision and audio capabilities are leading us to a simple conclusion - AI is moving ahead. They are able to do work, evaluate their own work and supervise themselves as hierarchal teams of AI agents. It is inevitable - imagine an ERP user whose primary work is to work on email, documents and ERP system as a backoffice user. Now imagine an AI agent that opens an ERP system, as access to email and documents with grounded RAG. Evertyhing this guy was doing - even creatively - can be replaced by this AI agent. Dont get trapped in old school thinking about AI - read the new paradigm in the book.