Generative AI Leader Practice Question: Techniques to Improve Generative AI Model Output
Which TWO techniques effectively reduce bias in generative model outputs? (Choose two.)
⚠ Common exam trap
A common misconception is that randomness (temperature) or model size alone can fix bias, when in fact these parameters do not address the systematic skew in training data or model representations.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Apply adversarial debiasing during training or fine-tuning.
Adversarial debiasing (A) directly reduces bias by training the model to minimize an adversary's ability to predict protected attributes from the model's outputs, forcing the model to learn representations that are invariant to those attributes. Fine-tuning on a balanced dataset (D) corrects representation bias by ensuring the model sees equal examples across groups, preventing overfitting to majority patterns. Both techniques actively address the root causes of bias in training data or model behavior.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Apply adversarial debiasing during training or fine-tuning.
Why this is correct
Adversarial debiasing trains a discriminator to detect protected-attribute predictions from the model's representations while the generator learns to defeat it, directly penalising biased internal representations during training or fine-tuning rather than only filtering outputs afterwards.
- ✗
Increase the temperature parameter to introduce more variability.
Why it's wrong here
Temperature controls sampling randomness, not the underlying data distribution, so raising it diversifies outputs while leaving biased associations intact. It is tempting because varied outputs appear less stereotyped, and it would be the correct choice when the goal is creative diversity rather than fairness.
- ✗
Use a larger model with more parameters.
Why it's wrong here
Scaling parameters does not remove bias encoded in training data or objectives; larger models can reproduce and amplify those patterns. It is tempting because capacity improves capability and reasoning, and would be the right lever for accuracy or task performance rather than fairness.
- ✓
Fine-tune on a dataset with balanced representation across groups.
Why this is correct
Balanced representation across groups in the fine-tuning dataset prevents the model from learning correlations skewed toward over-represented groups, directly reducing biased outputs at the data level rather than relying on post-hoc filtering or prompt engineering.
- ✗
Reduce max output tokens to limit the model's expression.
Why it's wrong here
Token limits truncate responses, which curtails expression without altering the model's learned associations or the prompts that elicit bias. It is tempting because shorter outputs expose fewer biased statements, and it would be correct when controlling cost, latency or verbosity.
Go deeper
Related to this question
About these practice questions
This Generative AI Leader question is part of Courseiva's 1,008-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This Generative AI Leader practice question is part of Courseiva's free Google Cloud certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the Generative AI Leader exam.