1. The AI4RE syllabus states that "not every AI is a chatbot". Which five technology families appear in almost every AI taxonomy it refers to?
2. What is the relationship between Large Language Models and natural language processing?
3. A Requirements Engineer must produce auditable, deterministic answers for a tax compliance check where no error rate is acceptable. Which AI family is the syllabus's stated fit for such domains?
4. In the chatbot processing pipeline, at which stage are the system context and the session or user context merged?
5. Why does a chat-based assistant re-read the entire conversation on every new message?
6. Which stage of the pipeline is described as the "run-time" phase, where the trained system applies what it learned during training to new input?
7. Which statements about fine-tuning are correct? Select all that apply.
8. Which two drawbacks does the syllabus attribute specifically to RAG?
9. A project needs an assistant that answers from a regulatory standard which is revised periodically, and every answer must be attributable to the clause it came from. Which approach fits and why?
10. A stakeholder reports that a constraint they stated at the start of a long chat session no longer influences the assistant's answers. What is the most likely explanation?