Google has paused the new “inclusive language” function it rolled out last month for its online word processor after facing heavy criticism online.
The controversial feature flagged potentially inappropriate language and prompted Google Docs users to replace them with alternative terms.
For example, the online word processor suggested replacing words such as “mankind” with “humankind,” “housewife” with “stay-at-home-spouse” and “landlord” with “property owner.”
Using inappropriate or non-inclusive terms would also prompt a pop-up message saying, “Inclusive warning. Some of these words may not be inclusive to all readers. Consider using different words.”
According to the tech giant, the decision was made to temporarily disable the “inclusive language” feature as they make further improvements to the AI-powered tool. The feature was no longer available on Google Docs, as of noon on Tuesday, April 26.
Google spokesperson Jenny Thomson told The Daily Wire that the feature is a “form of AI that uses language understanding models, based on millions of common phrases and sentences, to automatically learn how people communicate and suggest changes.”
A daily dose of Asian America's essential stories, in under 5 minutes.
Get our collection of Asian America's most essential stories to your inbox daily for free.
Unsure? Check out our Newsletter Archive.
“However,” she explained, “inclusive language suggestions—an assisted writing feature—can over or undercorrect certain phrases. We’re looking more carefully at the inclusive language suggestions and have paused those for further review while we continue to improve this feature.”
Aside from the “inclusive language” feature, Google also included AI-powered edits for better word choice, use of active voice and conciseness. The changes were made available to users with Business Standard, Business Plus, Enterprise Standard, Enterprise Plus and Education Plus accounts.
Despite Google’s aim to “elevate your writing style and make more dynamic, clear, inclusive, and concise documents,” the new features faced immediate backlash from users and advocates after they first started rolling out on March 31.
For Silkie Carlo, the director of nonprofit privacy campaigning organization Big Brother Watch, Google’s new AI-powered features “aren’t assistive, they’re deeply intrusive.”
“This speech-policing is profoundly clumsy, creepy and wrong, often reinforcing bias,” she added. “Invasive tech like this undermines privacy, freedom of expression and increasingly freedom of thought.”
Vice editors reportedly found inconsistencies in Google’s new features. A transcribed interview of former Ku Klux Klan leader David Duke, for example, did not receive any suggestions despite his use of the N-word and numerous references to hunting Black people.
According to a Google spokesperson, the new writing features are still undergoing improvements.
“Assisted writing uses language understanding models, which rely on millions of common phrases and sentences to automatically learn how people communicate,” the representative told Vice. “This also means they can reflect some human cognitive biases. Our technology is always improving, and we don’t yet (and may never) have a complete solution to identifying and mitigating all unwanted word associations and biases.”