Bias in AI and real-world consequences in AI for Everyone - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding Bias Sources in AI

Which of the following is the most common source of bias in AI systems?

A Biases present in the training data used to teach the AI

B Errors in the hardware components of AI devices

C The speed at which the AI processes information

D The programming language used to develop the AI

💡 Hint

Think about what teaches the AI how to make decisions.

Bias in AI often comes from the data it learns from. If the training data has unfair or unbalanced information, the AI will reflect those biases.

🚀 Application

intermediate

Real-World Impact of AI Bias

What is a likely real-world consequence if an AI system used for hiring favors one group over others?

A The AI will improve fairness by selecting the best candidates equally

B The AI will slow down the hiring process significantly

C Certain groups may be unfairly excluded from job opportunities

D All candidates will receive the same job offers regardless of qualifications

💡 Hint

Consider what happens if the AI prefers some people unfairly.

If the AI favors one group, it can lead to discrimination, where some people do not get fair chances for jobs.

🔍 Analysis

advanced

Identifying Bias Effects in AI Outputs

An AI system used for loan approvals consistently denies loans to applicants from a particular neighborhood. What is the most likely explanation?

A The training data included biased information linking that neighborhood to higher risk

B The applicants from that neighborhood have better credit scores

C The AI is programmed to randomly reject applications

D The AI is malfunctioning due to hardware failure

💡 Hint

Think about what the AI learned from before making decisions.

The AI likely learned from data that unfairly associates that neighborhood with risk, causing biased loan denials.

❓ Comparison

advanced

Comparing Bias Mitigation Strategies

Which approach is most effective in reducing bias in AI systems?

A Using a single source of data to keep consistency

B Increasing the AI's processing speed

C Limiting the AI's access to external information

D Using more diverse and representative training data

💡 Hint

Think about how the AI learns and what affects its fairness.

Diverse and representative data helps the AI learn fairly about all groups, reducing bias in its decisions.

❓ Reasoning

expert

Evaluating Long-Term Consequences of AI Bias

Consider an AI system used in criminal justice that has biased predictions against a minority group. What is a likely long-term societal consequence?

A Increased trust in the justice system by all communities

B Worsening social inequalities and reduced trust in institutions

C Faster case processing with no impact on fairness

D Equal treatment of all individuals regardless of background

💡 Hint

Think about how unfair treatment affects society over time.

Biased AI in justice can deepen inequalities and cause communities to lose faith in fairness and institutions.