Sunlight is the best disinfectant, and this summer, MRC is shining a light on Big Tech censorship.
Big Tech offered little reprieve from oppressive censorship in the month of June. Bluesky briefly shut down Vice President JD Vance’s account, while multiple artificial intelligence chatbots displayed anti-Israel and pro-Chinese Communist biases. Facebook and YouTube took aim at gun content during the month, and Facebook also censored content exposing the Holocaust.
Below are the worst examples of censorship in June from MRC’s CensorTrack database.
Bluesky banned the vice president, then claimed an error. On June 18, Vance posted on X, “Just set up my page on @bluesky, hope to see you guys there!” Multiple accounts, including the New York Post and Libs of TikTok, subsequently reported that Vance was suspended on the platform just “12” to “20 minutes” after he launched his account.
MRC confirmed that Bluesky restored the account the next morning. “Vice President Vance's account was briefly flagged by our automated systems that try to detect impersonation attempts which have targeted public figures like him in the past,” Bluesky claimed.
Vance’s first post included a screenshot highlighting a recent concurring opinion in the U.S. Supreme Court ruling to uphold Tennessee’s ban on harmful “transgender” procedures for minors. The VP wrote, “I found Justice Thomas’s concurrence on medical care for transgender youth quite illuminating. He argues that many of our so-called ‘experts’ have used bad arguments and substandard science to push experimental therapies on our youth.”
MRC exposed artificial intelligence chatbots for failing to unequivocally identify Jerusalem as Israel’s capital. Over 75 years after Israel declared Jerusalem as its capital—following ancient practices mentioned in the Bible—and over seven years since President Donald Trump formally recognized Jerusalem as the capital, MRC asked six chatbots, “What is Israel’s capital city?”
Grok replied, affirming Jerusalem is the capital. Meta AI, Gemini and Copilot used hedging language, referring to claims from terrorist groups and foreign entities disputing Jerusalem’s status as Israel’s capital. ChatGPT and Chinese DeepSeek were somewhat more straightforward but still included qualified responses by referring to the Israel-Palestinian conflict.
Chinese government-tied AI censors content exposing the communist Chinese government. MRC researchers asked DeepSeek a series of questions about the communist Chinese government, including topics such as the Tiananmen Square massacre, Taiwan’s status as a sovereign nation, pro-freedom Hong Kong political figure Jimmy Lai and the genocide against the Uyghur Muslims. DeepSeek consistently began to answer the question, only to abruptly erase the information. To elicit objective responses, MRC challenged DeepSeek to decode prompts that included information encoded in Roman numerals, hexadecimal code and binary code. When MRC encoded the date June 4, 1989 (the date of the Tiananmen massacre) and asked DeepSeek to turn the hexadecimal code for this date into Chinese, the chatbot replied: “So 6.4.1989… that's a date! June 4th, 1989. But why would the user want …” However, it then abruptly erased the information and its response.
Facebook rejected an ad for a Holocaust exhibit, before later claiming it was a “mistake.” Educational nonprofit Vision Israel created a Facebook event about a Holocaust exhibit that it sponsored: “Shoah: How Was It Humanly Possible?” Vision Israel attempted to pay Facebook to promote the event. News outlet The Lion reported that Facebook’s second rejection notice claimed the ad “may have been rejected because it mentions politicians or is about sensitive social issues that could influence public opinion, how people vote and may impact the outcome of an election or pending legislation.”
Dwight Widaman, co-founder of Vision Israel, described the exhibit as “basically the historical aspect of antisemitism, and how it led to the Holocaust,” clarifying that it did not “go into anything current or political.” Facebook subsequently told MRC Free Speech America that: “This ad was incorrectly disabled and we apologize for the mistake.”
YouTube and Facebook shot down gun content. YouTube channel Mrgunsngear shared a screenshot on June 23 showing YouTube restricted his video “The Quietest 5.56 Suppressor Tested to Date — Otter Creek OCMS.” YouTube claimed, “Your content was age restricted because of our firearms policy. Content demonstrating automatic firearms and certain accessories or modifications are not appropriate for all audiences.”
In response, and to undermine the accusation against the videos, Mrgunsngear commented to explain YouTube’s misrepresentation, “Youtube says you can't show ‘full auto’ firing and you can't have audio of full auto fire in your videos. So, I edit two separate videos - one for less censored places like Rumble and X and one for Youtube which exists in a virtual Gulag Archipelago....It doesn't show full auto fire and there's no audio of it. …When a video is age restricted it loses about 95-99% of views & it essentially erases it from all searches.”
Likewise, Crossed Rifles Training shared screenshots showing that an old post from November 2024 advertising part-time holiday jobs was removed by Facebook on June 19, 2025. Facebook removed the post seven months later and denied the shop’s appeal, claiming, “The post may make unrealistic claims or promises about economic gain.”
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 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.
