GMAT Verbal study materials
GMAT Verbal study resources

The Reading Comprehension Passage

GMAT Verbal · Put your knowledge to work in context.

All case studies

The Reading Comprehension Passage

Passage: 'In recent years, the use of artificial intelligence (AI) in hiring has grown rapidly. Proponents argue that AI reduces unconscious bias by focusing on job-relevant criteria. However, critics point out that AI systems are trained on historical data, which may contain biases. For example, if a company historically hired mostly from certain universities, the AI might learn to favor candidates from those universities. A study of 50 companies found that those using AI for initial screening had a 10% increase in diversity hires, but the study did not control for other factors such as changes in recruitment strategies. Furthermore, the study measured diversity only in terms of gender and ethnicity, not other dimensions such as socioeconomic background.'

1. What is the main point of the passage?

Read the worked answer

The passage presents a balanced view of AI in hiring. It notes that proponents believe AI reduces bias, while critics warn that AI can perpetuate bias from historical data. The passage then cites a study showing increased diversity hires at companies using AI, but points out methodological limitations: the study did not control for other factors and measured diversity narrowly. The main point is that the evidence on AI's effect on diversity is mixed and the study has limitations, so conclusions should be drawn cautiously.

2. Which of the following, if true, would most weaken the critics' argument?

Read the worked answer

The critics argue that AI systems trained on historical data may contain biases. To weaken this, we need evidence that AI systems do not necessarily perpetuate historical biases. For example, if AI systems are designed to ignore certain demographic information and are regularly audited for bias, and if studies show that AI screening actually increases diversity compared to human screening, that would weaken the critics. Specifically, the statement 'AI systems that are trained on data from diverse companies and are programmed to prioritize job-relevant skills have been shown to reduce bias compared to human reviewers' would weaken the critics' argument.