Google Removes 'Diversity' and 'Equity' from Responsible AI Page
Google Removes DEI Language from Responsible AI Team Webpage
Google has made subtle alterations to the webpage dedicated to its Responsible AI and Human Centered Technology (RAI-HCT) team. This team is specifically focused on research concerning AI safety, fairness, and explainability.
The updates involve the removal of terms like “diversity” and “equity” from the page’s content. Previously, the webpage utilized phrasing such as “marginalized communities,” “diverse,” “underrepresented groups,” and “equity” when outlining the team’s objectives.
Changes to Terminology
This language has now been either removed entirely or substituted with more general terms. For instance, “all,” “varied,” and “numerous” are now used in place of the former “diverse” descriptor.
A response from Google regarding these modifications was not immediately available when requested.
Broader Trend in Big Tech
These changes were initially observed by the watchdog organization, The Midas Project. They follow a similar removal of DEI-related language from Google’s Startups Founders Fund grant website.
Earlier in February, Google announced it would be discontinuing its diversity hiring targets and undertaking a review of its existing diversity, equity, and inclusion (DEI) programs.
Google is not alone in this shift. Numerous other large technology companies are also scaling back their DEI initiatives.
Increased Scrutiny and Policy Changes
- The current trend coincides with increased scrutiny from the Trump administration, which has labeled certain DEI practices as “illegal.”
- Amazon and Meta have also recently revised their DEI measures.
- OpenAI has removed references to diversity and inclusion from its hiring practices webpage.
- Conversely, Apple recently resisted a shareholder proposal aimed at ending its DEI programs.
It is noteworthy that many of these companies, including Google, maintain contractual agreements with various federal agencies.
