Agentic AI
Kartavya Desk Staff
Source: WEF
Subject: Science and Technology
Context: Agentic AI is rapidly gaining traction in the market as businesses adopt autonomous, goal-driven AI agents to automate complex workflows with minimal human oversight.
About Agentic AI:
What it is?
• Agentic AI refers to autonomous, goal-oriented AI systems capable of completing tasks independently with limited human oversight.
• It consists of AI agents—often powered by large language models (LLMs)—that can reason, make decisions and take actions in dynamic environments.
• In multi-agent systems, different agents handle specialised subtasks coordinated through AI orchestration.
How Agentic AI works?
• Perceives the environment: The system gathers real-time information from users, sensors, databases, APIs or the internet to understand the current situation.
• Understands and reasons: It analyses the collected data using language, vision or pattern recognition to interpret context and identify what needs to be done.
• Sets goals and plans: Based on user instructions or predefined objectives, the AI decides clear goals and plans a sequence of steps to achieve them.
• Decides and acts: The AI evaluates different possible actions, selects the best option and executes it by using tools, software, APIs or connected systems.
• Learns and coordinates: It reviews outcomes, learns from feedback and coordinates with other AI agents to improve future performance and complete tasks efficiently.
Key features
• Autonomy: Performs multi-step tasks without continuous human intervention.
• Proactivity: Can initiate actions, monitor systems and respond to changing conditions.
• Tool-use capability: Interacts with external tools, databases and applications.
• Specialisation: Agents can be task-specific or organised hierarchically or horizontally.
• Adaptability: Learns from experience and improves performance over time.
• Natural interaction: Operates through natural language, reducing dependence on complex user interfaces.
Significance:
• Enables end-to-end automation of complex workflows beyond simple content generation.
• Boosts productivity and efficiency by reducing human cognitive and operational load.
• Supports advanced applications in enterprise operations, software development, robotics, healthcare, finance and logistics.