I’ve been thinking about Agentic AI.
Essentially it’s similar to other AI but with a focused knowledge area, capable of making its own decisions based on the knowledge it has and the ability to learn and reason without human intervention. It gathers data, processes it, makes decisions based on that data and its assumptions, then learns from feedback. Try try and try again and mae changes based on your failures.
It is generally tasked towards a specific function such as event planning, task scheduling, predicting something, etc.
It’s used in self-driving cars, supply chain management, cybersecurity, healthcare, financial services, and anything that can be automated.
Concerns range from data handling, privacy, to replacing humans.