OutSystems Generative AI/Agent/Agentic AI
Question: Which of the following best describes a key concept of Generative AI?
A. Generative AI systems are designed only to classify existing data without creating new outputs.
B. Generative AI models can produce new content (text, images, audio, code) by learning patterns from large datasets.
C. Generative AI relies exclusively on rule‑based programming and does not use machine learning.
D. Generative AI cannot adapt to different domains or tasks.
Correct Answer: B – Generative AI models generate new outputs by learning from patterns in training data, enabling applications like text generation, image synthesis, and code completion
Question: Which of the following best explains how Generative AI is applied in OutSystems?
A. It only classifies existing app data without creating new outputs.
B. It generates new content such as code snippets, workflows, or UI layouts by learning patterns from large datasets.
C. It relies exclusively on rule‑based programming and does not use machine learning.
D. It cannot adapt to different domains or app requirements.
Correct Answer: B – In OutSystems, Generative AI can auto‑generate code, UI components, and documentation by learning from patterns, accelerating low‑code development and enhancing productivity.
Question: Which statement best reflects the core design principle of Agentic AI in OutSystems?
A. Agentic AI agents only respond to prompts and cannot act autonomously.
B. Agentic AI agents are designed to reason, plan, and take independent actions toward goals, while being governed through the OutSystems Agent Workbench.
C. Agentic AI relies exclusively on static rule‑based programming without adaptability.
D. Agentic AI cannot collaborate with humans or other agents in enterprise workflows.
Correct Answer: B – Agentic AI in OutSystems emphasizes autonomous reasoning and planning, with the Agent Workbench providing governance, lifecycle management, and secure integration into enterprise systems.
Question: Which statement accurately describes the difference in task execution between Single-Agent and Multi-Agent workflows?
A) Single-Agent workflows achieve higher scalability for complex tasks by using a single, powerful reasoning model.
B) Multi-Agent workflows naturally require simpler governance and fewer specialized observability and tracing tools.
C) Single-Agent workflows have lower fault tolerance because the entire task workflow represents a single point of failure.
D) Multi-Agent workflows exhibit high modularity, allowing agents to be developed, tested, and replaced independently.
Correct answer: D) Multi-Agent workflows exhibit high modularity, allowing agents to be developed, tested, and replaced independently.
Question: In the context of building an Agentic App on OutSystems Developer Cloud (ODC), why would an architect integrate an external AI Search service (such as Azure AI Search or Amazon Kendra) into the application’s design?
A. To replace the OutSystems runtime engine with a third‑party service.
B. To enable the agent to retrieve, index, and reason over enterprise knowledge sources efficiently.
C. To bypass governance and security controls in OutSystems.
D. To eliminate the need for human oversight in agent workflows.
Correct Answer: B – External AI Search services provide scalable indexing and retrieval of enterprise data, allowing Agentic Apps to access relevant knowledge in real time while OutSystems ensures governance and orchestration.
Question: In designing an AI Agent, how does the is executed or skipped?
Answer (explained in the image above):
- Condition TRUE → Action Executes (the agent triggers the defined action).
- Condition FALSE → Action Skipped (the agent bypasses the action).
The infographic I generated shows this logic flow clearly:
- Central AI Agent branching into two paths.
- Green path (TRUE): Action is triggered and executed.
- Orange path (FALSE): Action is ignored/skipped.