Since reading a survey on Agentic AI, I've formulated a basic blueprint of what I feel it is.
Agentic AI basically concerns creating AI workers that are:
- Autonomous, Adaptable and goal-driven
- Use a combination of reinforcement learning (RL) and goal-oriented architectures/approaches
- Implements adaptive control strategies and techniques
- Designed to be resilient to change (and the unknown)
- Implements smart, opportunistic learning such as RL, imitation, pattern knowledge, priorities, social interactions, self-supervision, uncertainty, real-time and dynamic and real-time datasets and environments.
- Long-term management of strategies concerning goals, tasks, context, priority and any other smarts such as patterns, cause and effect analysis, focus and attention.
- Management of complexity and overload
- Modular, flexible, composable, combinable, generalizable and evolvable concepts.
- Redefinable concepts of value
- Drawing from knowledge and historical events (cause and effect)
- Teaching methods
- Training/Learning methods, transfer learning
- Self-reflection (Performance, Quality, Goals, Learning)
- Safety and security, and ethical decisions
- Characterising a situation (defining its goals, priorities, subjects, interactions, attribution of cause and effects, etc)
- Monitoring
- Hardware and resources, and scalability
- Understanding changes in human behaviour (i.e the changing of human goals, etc.)
This makes Agentic AI very commensurate with simulating NPC behaviour in real-time environments such as computer games.