3.1 Explain Responsible AI Principles (fairness, Transparency, Accountability, Bias Mitigation, Safety) AI-TECHNICAL-PRACTITIONER Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   What is a strategy to mitigate bias in AI?

   • AIncreasing the model's complexity to capture more patterns
   • BDeploying the model without testing
   • CRemoving all features related to protected attributes
   • DUsing diverse and representative training data
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A hospital uses an AI system to assist in diagnosing diseases. A misdiagnosis occurs, and the hospital needs to determine who is responsible. Which practice would have best supported accountability?

   • AUse a black-box model to avoid scrutiny of the decision process.
   • BEnsure the AI system is highly accurate, assuming errors will not occur.
   • CMaintain detailed logs of the AI system's inputs, outputs, and decision-making process.
   • DRely on the AI vendor's warranty to cover any errors.
   Reveal Solution Discussion Select an answer first
4. 8foundation · easy

   Which of the following is an example of an accountability mechanism in AI?

   • AConducting regular audits of AI systems to ensure they operate as intended
   • BDeploying AI systems as quickly as possible to gain a competitive edge
   • CTraining AI models on large datasets
   • DUsing the most advanced AI algorithms available
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   Why is responsible AI important for Cisco AI solutions?

   • AIt is required by law in all countries where Cisco operates
   • BIt ensures that AI systems are built with the latest technology to outperform competitors
   • CIt helps build trust with customers and stakeholders by addressing ethical, legal, and societal implications
   • DIt guarantees that AI models will be 100% accurate in all scenarios
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   An e-commerce company is building a recommendation system. The training data contains user behavior logs that may reflect existing biases in product exposure. The company wants to mitigate bias before deployment. Which action is most effective?

   • ACollect additional data from underrepresented user groups to balance the dataset.
   • BDeploy the model and monitor for bias after it is in production.
   • CRemove all user demographic information from the dataset.
   • DUse a more powerful machine learning algorithm to improve overall accuracy.
   Reveal Solution Discussion Select an answer first