Generative AI

60 Minutes

36 Questions

This test is designed to assess a candidate's knowledge of Generative AI concepts and fundamentals. It will include multiple-choice questions on topics such as large language models (LLMs), prompt engineering techniques, model training and inference basics, and common applications of generative AI. The assessment will also cover key challenges such as hallucinations (factually incorrect or fabricated outputs), bias, and limitations of current models, along with strategies to mitigate these risks.

In addition, candidates will be tested on responsible AI usage, handling sensitive data, privacy and compliance considerations, ethical implications, and evaluating AI outputs for accuracy and reliability. The test is designed to evaluate the candidate's ability to understand and apply core concepts of generative AI, while recognizing both its potential and its risks in practical scenarios.

Example Question:

Multiple-Choice

You are responsible for reviewing a proposal to use GenAI to generate personalized career development advice for employees. The system would base suggestions on their current roles, past reviews, and demographic information. Which ethical risk is most critical to address?

Answers

1. The system may generate advice that is too detailed to implement.

2. Employees may ignore the recommendations if the tone is too formal.

3. The system could reinforce existing biases or stereotypes based on demographic attributes.

4. The system may recommend irrelevant roles outside the company's needs.