YouTube Suddenly Removed 50 Million Videos Over Some Vague Policies

YouTube Suddenly Removed 50 Million Videos Over Some Vague Policies

YouTube, the popular video-sharing platform, has often made a song and dance about how their content ecosystem is near unmanageable. Cristos Goodrow, the site’s vice president of engineering, once wrote that “[YouTube] represents users rewarding their curiosity” through the 5.97 billion hours of content viewed through the platform every day, according to external stats.

Though for as long as the site has existed, YouTube has shown an eagerness to play the unaccountable judge, jury and executioner for the online market’s freedom of speech. As administrators display a will to selectively bend the rules, especially when it quells the pressure from governmental and internal forces, their users are routinely left without evidence, defensive representation, or a neutral arbitrator to judge the sudden banishment as just.

Their latest report showcases the extent of their enforcement. Earlier this week, YouTube published their quarterly report alleging how many videos the site was able to remove as “problem content” deemed to be a violation of their terms of services policies. Between July and September, we’re lead to believe YouTube took down over 50.2 million videos, nearly 1.7 million channels and over 224 million comments. These cases were reportedly handled through their 10,000 random contractors deemed “trusted flaggers”, NGOs, government agencies and A.I. algorithms.

How does this “problem content” break down? Well, it’s unclear given their report isn’t entirely congruent. If we’re to believe the word of YouTube, nearly 75% of the removed videos were spam, 10% were specifically violations of hate speech, adult content and child safety (despite CEO Susan Wojcicki insisting the site isn’t for minors) and only a staggering 0.4% showcasing violence or extremism.

If we go by the report, however, these statistics are incredibly smaller. On paper, the company states only 7.8 million videos were removed during the last three months, meanwhile, their channel and comment removal stats appear consistent. This means they’re only breaking down a fraction of a fraction of their overall removed video content. Are we seriously dealing with typos from the world’s most powerful search engine, making technical oopsies with their overall censorship efforts, or are we missing all their appropriate data?

The report defines none of the offences, mentions no targeted individuals, provides no access to a records database that would explain the context behind each ban (which the company probably doesn’t have) and fails to offer transparency on the site’s way of anarcho-capitalist justice. It seems only Engadget’s tech journalist Mallory Locklear is noting the report’s admission 81% of the content was judged by the company’s automated systems and 74.5% didn’t receive a single view before being censored.

Where’s the due process on behalf of YouTube? Outsourced to reactionaries?

“We’ve always used a mix of human reviewers and technology to address violative content on our platform,” the company said in a recent statement. “In 2017 we started applying more advanced machine learning technology to flag content for review by our teams. This combination of smart detection technology and highly-trained human reviewers has enabled us to consistently enforce our policies with increasing speed.”

So does speed take priority over justice? And does YouTube truly care about a transparent, ethical process before taking away online liberties anyway? Again, if we take the company at their word, of course, they do. Admittedly, channel terminations do incur a more sophisticated process than videos and comments. Users have their platform removed only after they’ve accumulated three strikes for violating community guidelines displaying “severe abuse” or demonstrate that they are “wholly dedicated” to violating the rules.

The appeals process, however, is rigged in the company’s favor. Their content policies can be subject to change at any time, can be attributed against previously approved videos, can be served all at once and can only be disputed and exonerated by YouTube themselves, not an independent body with qualifications outside of technical know-how. Should their corporate whims disagree with your defense, justice always goes their way.

This isn’t to say the statistics, while unverifiable and vague, couldn’t have some truth behind them. YouTube is certainly no perfect platform where the censored are all innocent dissidents. The controversial Alternative Influence Study from Data & Society, while using flawed methodologies and showing obvious biases towards censorship, showcased the web of online personalities that can actually lead one from ethnic left-wing comedian Bunty King to classical liberal pundit Sargon Of Akkad to race realist Stefan Molyneux to actual members of the alt-right apologizing for violence, segregation, fascist conspiracies and white nationalism.

Their use of snowball sampling, without providing necessary qualitative research examining whether extremist content was negative or positive, did show YouTube’s problem of fostering an online web towards extremists often left untouched. The platform routinely enables harassment, the spread of propaganda from neo-Nazi and Islamist groups and ridiculous, defamatory conspiracy theories claiming large numbers of politicians and celebrities molested children in a pizzeria.

The issue remains whether they’ve banished videos without merit, if the company has found someone committing a legitimate violation, and if substantial evidence has been presented before an independent arbitrator and a decision was made ethically. As stated by Vox’s editor Ezra Klein, people shouldn’t just scroll through “one of the must underestimated forces in politics right now.” This struggle between moral administration and freedom of speech, which doesn’t impede on the liberties of others, should be at the forefront of YouTube’s priorities if it wants to reign as the tech monopoly where billions exercise their online rights. This starts with clearly defined codes of conduct, impartial judgement and ends with a commitment to these fundamental rights before easy PR and profits.