According to a report by MIT Technology Review AI, artificial intelligence could soon begin screening job applicants' resumes before any human sees them. However, there are serious grounds to doubt these systems' ability to evaluate candidates impartially.
Researchers have already established that large language models inherit human biases directly from the data on which they are trained. This means algorithms may reproduce and even amplify existing societal stereotypes when making hiring decisions.
This situation creates a risk that automated screening will prove to be a less objective tool than traditional human selection, despite the widespread belief in the technical neutrality of machines.
editorial commentary
Why it matters
The likely consequence will be increased regulatory pressure on companies using algorithmic hiring, demanding audits of training samples. The next observable signal will be lawsuits or official investigations into cases of discrimination via AI. The key uncertainty lies in the speed with which the market can develop reliable standards for cleansing data of social biases.