The following text discusses algorithmic bias.
Artificial intelligence systems can perpetuate and amplify societal biases. Facial recognition algorithms trained predominantly on light-skinned faces show higher error rates for darker skin tones. Hiring algorithms have reproduced historical discrimination patterns found in training data. Researchers are developing techniques to audit and mitigate these biases, though no perfect solution yet exists.
What is the main purpose of the text?
To argue that AI should be banned entirely
To explain how AI bias occurs and note efforts to address it
To compare different AI companies' ethics policies
To provide technical instructions for building fair algorithms
Correct Answer: B
Choice B is the best answer. The text explains how bias occurs (training data, error rates) and notes efforts to address it (audit and mitigation techniques).
- Evidence: The text explains mechanisms: "trained predominantly on light-skinned faces," "reproduced historical discrimination." It notes efforts: "Researchers are developing techniques to audit and mitigate."
- Reasoning: The passage identifies a technical/social problem and describes the current state of solutions.
- Conclusion: The purpose is to explain the bias and efforts to address it.
💡 Strategy: Summarize: Problem (Bias) + Solution (Audit/Mitigation).
Choice A is incorrect because banning isn't advocated. Choice C is incorrect because companies aren't compared. Choice D is incorrect because technical instructions aren't given.