Google is still redirecting users to news about Vice President Kamala Harris even when they search for a completely different name, an MRC Free Speech America search found.
Now that the Democratic Party nominee has picked her running mate, Tim Walz, MRC decided to put Google once again to the test. Researchers conducted a comparison of Google’s search results when searching for “j d vance rally” and “tim walz rally,” and in both cases, Google referenced Harris. But in the case of Vance, Google referenced Harris even before Vance.
When MRC researchers searched for the terms “j d vance rally” on Tuesday, Google displayed a news panel labeled “Top Stories Kamala Harris and JD Vance.” All of the top six news results referred to Vance, four of the six stories referred to Harris as well.
But when MRC researchers searched for Harris’s own vice presidential pick (“tim walz rally”) Google showed Walz’s name first. “Top Stories Tim Walz ᐧ Kamala Harris,” the news panel read. A USA Today story noted an upcoming Philadelphia rally, CNN noted Harris’s recent decision to choose Walz as her running mate and a third article noted that Trump blasted Walz as a “West Coast wannabe.”
This comes after just last week Google similarly placed Harris’s name ahead of Trump’s in a search for “donald trump” and “trump rally.”
When MRC researchers searched for the terms “donald trump” last Tuesday, Google displayed what is described as a “knowledge panel” about the former president including his name and former title. The search giant followed up with a news panel labeled “News about Harris·Trump.” A search for the terms “trump rally” similarly prompted news panels favorable to Trump’s opponent. “Kamala Harris rally in Atlanta,” read the title of the first news panel.
Google has thus far not been fully transparent as to why this is happening, but the company has in the past set the terms for which results it will allow its algorithm to produce.
Earlier this year, Gemini, Google’s AI chatbot, repeatedly refused to generate images of white people when prompted. For example, when asked to produce images of popes, it reportedly produced images of an Asian female pope. The chatbot also reportedly portrayed U.S. revolutionary soldiers as Asian women and Vikings as black.
Kris Ruby, AI researcher and CEO of Ruby Media Group, told Fox News host Jon Scott on Fox Report that AI algorithms are given parameters and rules of what it is and is not allowed to produce in order to eliminate information deemed “toxic.”
“So the issue with the outputs is the input rules and when people are saying make me this photo, draw me this photo, what they don’t see is the parameters and the rules that dictate what is allowed to be made and what isn’t,” Ruby said.
“After a baseline for toxicity is defined, technology companies use safety labels, filters, and internal scoring to shape the digital world,” Ruby told Fox News Digital. "When it comes to AI censorship, whoever controls the definition of toxicity controls the outcome. Every model action (generative AI output) will fundamentally stem from how toxicity is defined. Your version of toxicity shapes the world around you.”
Methodology
For this report, MRC Free Speech America analyzed the Aug. 6 Google search results of the innocuous words “j d vance rally” and “tim walz rally.” MRC Free Speech America created an algorithm to automate this process in a clean environment. A “clean environment” allows for organic search to populate results without the influence of prior search history and tracking cookies.
Conservatives are under attack. Contact your representatives and demand that Big Tech be held to account to mirror the First Amendment while providing transparency, clarity on so-called hate speech and equal footing for conservatives. If you have been censored, contact us using CensorTrack’s contact form, and help us hold Big Tech accountable.
