Amazon has reportedly scrapped a recruitment tool powered by artificial intelligence (AI) that showed a clear bias against women.
“Everyone wanted this holy grail”, one of the sources said. The system was designed to sift through 100 job applications and “spit out” the top five results.
The Week
Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.
Sign up for The Week's Free Newsletters
From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.
From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.
The machine gave each applicant a rating out of five, “much like shoppers rate products on Amazon”, says the news site.
The AI was trained to use a method known as machine learning. This allowed researchers to feed the program old CVs from the past ten years to use as reference material so the AI could accurately grade real-life job applications, the BBC says.
But the data submitted came primarily from former male applicants, meaning the “sexist” AI taught itself that “male candidates were preferable”, the broadcaster reports.
The bias was discovered in 2015 when recruiters found that the AI system wasn’t rating applicants in a gender-neutral way. The program was penalising CVs with the word “women” in them, putting female applicants who had studied at women’s colleges, for example, or written “women’s chess club captain” in their list of achievements at a disadvantage.
Programmers subsequently amended the system to remove the bias but this was not a complete success, according to Fortune. The AI was abandoned last year.
Amazon declined to comment on the technology, but said the tool “was never used by Amazon recruiters to evaluate candidates.”
Has the AI hindered female job applicants?
Amazon insiders insist it hasn’t.
Speaking to The Sun, Amazon sources said the retail giant “never based recruitment decisions on the tool's ratings alone”, but recruiters did examine the AI’s recommendations when sifting through applications.
This isn’t the first time an AI programme has been shown to favour men over women.
Facebook came under fire last month after it emerged that AI systems had been preventing women from seeing certain job advertisements, the newspaper says.
-
Claude Code: the viral AI coding app making a splash in techThe Explainer Engineers and non-coders alike are helping the app go viral
-
‘Human trafficking isn’t something that happens “somewhere else”’Instant Opinion Opinion, comment and editorials of the day
-
What would a credit card rate cap mean for you?the explainer President Donald Trump has floated the possibility of a one-year rate cap
-
Claude Code: the viral AI coding app making a splash in tech
The Explainer Engineers and non-coders alike are helping the app go viral
-
Will regulators put a stop to Grok’s deepfake porn images of real people?Today’s Big Question Users command AI chatbot to undress pictures of women and children
-
Most data centers are being built in the wrong climateThe explainer Data centers require substantial water and energy. But certain locations are more strained than others, mainly due to rising temperatures.
-
The dark side of how kids are using AIUnder the Radar Chatbots have become places where children ‘talk about violence, explore romantic or sexual roleplay, and seek advice when no adult is watching’
-
Why 2025 was a pivotal year for AITalking Point The ‘hype’ and ‘hopes’ around artificial intelligence are ‘like nothing the world has seen before’
-
AI griefbots create a computerized afterlifeUnder the Radar Some say the machines help people mourn; others are skeptical
-
The robot revolutionFeature Advances in tech and AI are producing android machine workers. What will that mean for humans?
-
Separating the real from the fake: tips for spotting AI slopThe Week Recommends Advanced AI may have made slop videos harder to spot, but experts say it’s still possible to detect them
