The primary distinction between generative AI and agentic AI concerns the degree of autonomy, task execution capability, and workflow orchestration. Option C is the best answer because it contrasts conventional prompt-response generation with agentic systems designed to execute multi-step decision and action cycles.
Generative AI models primarily produce content such as text, images, audio, software code, or structured data in response to input.
Agentic AI extends these capabilities by incorporating goal-directed planning, external tool invocation, environmental feedback, and iterative task execution.
For example, an agentic network operations assistant might investigate a service outage, retrieve device telemetry, identify probable causes, recommend remediation, and request authorization before executing changes.
Option A is incorrect because both conventional generative AI applications and agentic systems can use RAG and external knowledge sources.
Option B is incorrect because agentic systems are specifically designed to perform authorized actions without requiring a new user prompt at every intermediate step.
Option D is incorrect because generative models do not inherently require tools, while agentic applications frequently depend on external tools and APIs.
A technical qualification is important: generative AI is not limited to single-turn conversations. It can support multi-turn interactions. The meaningful distinction is autonomous orchestration and action-oriented execution.
Specific Cisco AITECH Course Guide References:
Generative AI Ecosystem
AI-Powered Workflows and Agentic AI
Exam Objective 6.1: Differentiate agentic AI from generative AI use cases
Exam Objective 6.2: Autonomous capabilities and orchestration
Cisco identifies agentic autonomy and workflow execution as central learning outcomes.
===================
Submit