From: jimruttshow8596
Zachary Vorhies, a former software engineer who spent eight years at Google, has come forward with “provocative claims” about the company’s alleged improper and illegal actions regarding search results and blacklists [00:00:20]. He has publicly released a “large repository” of internal Google documents to support his claims [00:00:37].
Vorhies’ Allegations
Vorhies claims Google has been “lying to the American public about its meddling” in search results, YouTube queries, and elections [00:01:54]. He asserts that Google executives perjured themselves in congressional testimony by stating they do not employ blacklists [00:02:08]. Vorhies states that internal searches for “blacklist” at Google would yield results, including a popular YouTube query blacklist [00:02:26]. His decision to disclose this information was to inform the American public [00:02:42].
Google’s Response to the Whistleblower
Google attempted to stop Vorhies from releasing the documents [00:02:50]. Once they identified him as the leaker, they initiated a “wellness check” on him [00:02:56]. A wellness check is a tip called into the police, asserting concern for a person’s well-being or potential harm, which then gives police probable cause to search for and question the individual [00:03:16]. Vorhies believes Google used a “Deadman switch” he posted on Twitter as justification for the wellness check [00:04:45].
Observations on Google’s Internal Culture
Vorhies states that after the 2016 US presidential election, he observed a shift in Google’s internal culture [00:06:00]. During a weekly executive meeting (TGIF), executives expressed distress over Donald Trump’s election, attributing it to Russian hacking, racism, hate, and misogyny [00:07:01].
Around December 2016, “odd papers” began appearing on Google’s internal networks [00:07:58]. One such paper, titled “psychological induced stress as a result of algorithmic discrimination,” discussed how to alleviate user stress caused by perceived algorithmic discrimination [00:08:09]. It suggested providing “non-functional buttons and dials” to pretend to turn off censorship, thereby restoring a user’s “sense of control” and reducing stress [00:08:39]. Another part of the paper discussed simulating “crowding” to induce people to buy more products from ads [00:09:01].
Defining and Suppressing “Fake News”
According to Vorhies, Google executives began defining “fake news” as a reason for Trump’s election, asserting a need to protect the American public from it [00:10:25]. He states that internal papers on fake news cited examples of factual events, such as allegations concerning Hillary Clinton and weapons shipments, as “fake news” that needed to be censored [00:10:48]. Vorhies claims Google was “sculpting the information landscape” and filtering information [00:11:23].
Machine Learning Fairness (ML Fairness)
Google is allegedly building a system called “machine learning fairness” (ML Fairness) [00:12:15]. This system is described as “revising all of their search results,” “biasing their YouTube results,” and “revising all of their news” [00:12:48]. ML Fairness takes training data as input to generate rules applied to content to rank it based on “how fair it is” [00:13:05].
A document defined “algorithmic unfairness” with an example: if a search for CEOs predominantly returned male results, “even if this represented objective reality,” it could still be classified as “algorithmically unfair,” warranting “product intervention” [00:13:25]. Vorhies interprets this as Google trying to “change the nature of reality to make a better person” [00:14:10]. The documents also suggest that users are “programmed based upon the content that they’re interacting with” [00:14:29].
Vorhies’ core claim is that Google believes it is programming its users through content delivery and has a “moral obligation” or “business obligation” to program people to be “better” in ways Google defines [00:16:19]. He contends that Google has abandoned its stated objective stance of not manipulating information [00:17:03].
Allegations of Perjury and Political Bias
Vorhies claims that Google executives have perjured themselves before Congress [00:25:00]. He cites a Google representative, Karan Bhatia, testifying under oath that Google does not employ blacklists, despite Vorhies disclosing one and Robert Epstein allegedly discovering nine others [00:25:26]. He also claims similar statements from Google’s CEO regarding political censorship are false [00:26:09].
Vorhies provides an example of alleged political bias in Google search auto-prediction [00:26:24]: searching “Trump body-count” yields auto-predictions, while “Hillary body” (referring to a “horrendous body count list” he believes exists around Hillary Clinton) does not provide suggestions [00:26:26]. He suggests cross-referencing this with Google Trends, which he claims shows higher search traffic for “Hillary Clinton body count” compared to “Trump body count,” yet the auto-prediction is absent for Clinton [00:27:02]. Vorhies states that “the search terms that they’re Auto predicting are don’t reflect what users are actually searching for” [00:27:52], and that this alleged meddling is “all related to the political ideology” [00:28:06].
Alleged Election Interference
Vorhies claims Google has interfered in elections in the US, Ireland, and Brazil [00:28:31]. In Ireland, he alleges interference through the banning of the phrase “eighth amendment to the constitution of Ireland” in YouTube query blacklists [00:28:37]. In Brazil, he claims the Supreme Court ruled Google interfered in their election, citing a contract between Google and an opponent of Brazil’s current president [00:28:51]. These claims relate to the broader allegations of Google’s election interference.
Exploitable Vulnerabilities and Account Takedowns
Vorhies describes a “security vulnerability” within Google that a group of insiders allegedly exploit to take down accounts [00:40:01]. He explains that by creating a network of email accounts (e.g., xxx@gmail.com, xxy@gmail.com) and pushing spam, Google’s AI will deactivate the spam network and, “surprisingly,” also deactivate the original target email address [00:40:42]. He claims this pattern was used to take down Jordan B. Peterson’s email account and believes it was used for Tulsi Gabbard’s ad account suspension after a Democratic debate [00:41:59].
The Breaking Point: Translation Manipulation
Vorhies’ “breaking point” was discovering Google allegedly deleting translation words from the Arabic-to-English dictionary to make a Trump tweet “sound crazy” [00:43:30]. This alleged manipulation was then used by the New York Times to try to invoke the 25th Amendment to remove the President [00:43:40]. He cites the word “kaff” (kaf FA) as an example, which translates to “we will stand up” [00:43:54]. Google allegedly deleted the word twice because its AI was smart enough to find an alternative phonetic transliteration [00:45:10]. Vorhies viewed this as “treasonous behavior” and a national security issue, leading him to disclose the documents [00:45:46].
Call for Investigation and Whistleblowing
Vorhies believes that despite attempts to scrub records, evidence of these behaviors exists in Google’s vast archives [00:46:12]. He advises other Google employees with evidence of misdeeds to contact Project Veritas, emphasizing their “unique capability and integrity to maintain the discretion of their whistleblowers and to protect their identity” [00:47:19]. He specifically recommends using a VPN and a Tor browser for secure communication [00:48:32].
Vorhies’ comprehensive archive of internal Google documents is publicly available at projectveritas.com/google-dump [00:17:29], specifically the everything.zip file [00:17:36]. He encourages people to examine the documents themselves, stating, “don’t my word for it look at the documents that I’ve disposed that’s for all the credibility” [00:49:12].