EFF Highlights Stories Of Bad Content Moderation With New TOSsed Out Site

from the content-moderation-is-impossible dept

We’ve pointed out for many years that content moderation at scale isn’t just hard, it’s impossible to do well. At the scale of giant platforms, there needs to be some level of moderation or the platforms and users will get overwhelmed with spam or abuse. But at that scale, there will be a ton of mistakes — both type I and type II errors (blocking content that shouldn’t be blocked and failing to block content that probably should be blocked). Some — frankly dishonest — people have used a few examples of certain content moderation choices to falsely claim that there is “anti-conservative bias” in content moderation choices. We’ve pointed out time and time again why the evidence doesn’t support this, though many people insist it’s true (and I’ll predict they’ll say so again in the comments, but when asked for evidence, they will fail to present any).

That’s not to say that the big platforms and their content moderation practices are done well. As we noted at the very beginning, that’s an impossible request. And it’s important to document the mistakes. First, it helps get those mistakes corrected. Second, while it will still be impossible for the platforms to moderate well, they can still get better and make fewer errors. Third, it can help people understand that errors are not because someone hates you or has animus towards a political group or political belief, but because they fuck up the moderation choices all the time. Fourth, it can actually help to find what actual patterns there are in these mistakes, rather than relying on moral panics. To that end, it’s cool to see that the EFF has launched a new site, creatively dubbed TOSsed Out to help track stories of bad content moderation practices.

Just looking through the stories already there should show you that bad content moderation choices certainly aren’t limited to “conservatives,” but certainly do seem to end up impacting actually marginalized groups:

EFF is launching TOSsed Out with several examples of TOS enforcement gone wrong, and invites visitors to the site to submit more. In one example, a reverend couldn?t initially promote a Black Lives Matter-themed concert on Facebook, eventually discovering that using the words ?Black Lives Matter? required additional review. Other examples include queer sex education videos being removed and automated filters on Tumblr flagging a law professor?s black and white drawings of design patents as adult content. Political speech is also impacted; one case highlights the removal of a parody account lampooning presidential candidate Beto O?Rourke.

?The current debates and complaints too often center on people with huge followings getting kicked off of social media because of their political ideologies. This threatens to miss the bigger problem. TOS enforcement by corporate gatekeepers far more often hits people without the resources and networks to fight back to regain their voice online,? said EFF Policy Analyst Katharine Trendacosta. ?Platforms over-filter in response to pressure to weed out objectionable content, and a broad range of people at the margins are paying the price. With TOSsed Out, we seek to put pressure on those platforms to take a closer look at who is being actually hurt by their speech moderation rules, instead of just responding to the headline of the day.?

As the EFF notes, this is something of a reaction to the White House’s intellectually dishonest campaign database building program to get people to report stories of social media bias against conservatives. Unlike the White House, which is focused on pulling some one-sided anecdotes it can misrepresent for political points, the TOSsed out effort is a real opportunity to track what kinds of mistakes happen in content moderation and how to deal with them.

Filed Under: , , , ,
Companies: eff

Rate this comment as insightful
Rate this comment as funny
You have rated this comment as insightful
You have rated this comment as funny
Flag this comment as abusive/trolling/spam
You have flagged this comment
The first word has already been claimed
The last word has already been claimed
Insightful Lightbulb icon Funny Laughing icon Abusive/trolling/spam Flag icon Insightful badge Lightbulb icon Funny badge Laughing icon Comments icon

Comments on “EFF Highlights Stories Of Bad Content Moderation With New TOSsed Out Site”

Subscribe: RSS Leave a comment
Stephen T. Stone (profile) says:

bad content moderation choices certainly aren’t limited to "conservatives," but certainly do seem to end up impacting actually marginalized groups

…which is always the way it goes. LGBT content is often one of the first targets of sex-related content moderation because of unfair cultural assumptions that such content is sexual in nature. A picture of a same sex couple kissing is far more likely to get dinged than a picture of an opposite sex couple kissing, even if both pictures are not even remotely sexual.

TheLizard (profile) says:

Re: Fact-check: FALSE

Let’s clarify this a bit:

It’s not really "anti-conservative" bias, that’s easy to explain away, especially when you can define "conservative" one way or another.

It’s really more "anti-establishment" bias. Bias against the narrative of the global corporate and media elite.

That’s especially clear on Twitter, when blue-checked journalists are given a pass on obviously TOS-violating behavior, but any far-left or far-right viewpoint, or even populist ones, will get banned outright when they have a popular message that goes against the advertiser- and mainstream-journalist-friendly narrative.

Mike Masnick (profile) says:

Re: Re: Fact-check: FALSE

That’s especially clear on Twitter, when blue-checked journalists are given a pass on obviously TOS-violating behavior, but any far-left or far-right viewpoint, or even populist ones, will get banned outright when they have a popular message that goes against the advertiser- and mainstream-journalist-friendly narrative.

Two assertions are made here without any evidence. Can you back up either one and show that it is because of this bias, rather than — as clearly stated in the article — due to the vagaries and everyday errors that will come with content moderation?

If not, well, you failed.

TheLizard (profile) says:

Re: Re: Re:3 Fact-check: FALSE

I DID, you can’t be bothered to look into it.

That’s why all this banning WORKS. Because nobody SEES it, and people like you dismiss it because you can’t be bother to do any research yourself.

Here you go:


Next, you’ll post an excuse about how that kind of thing isn’t evidence.

I’ll post another example, and you’ll do the same.

After several, you’ll go, "well it’s TOS, they would do that to anybody," or "well they’re going to make mistakes."

You’ve made up your mind, social media reinforces your opinion, and on it goes until you don’t really have any opinions of your own, they are given to you instead.

Anonymous Coward says:

Re: Re: Re:4 Fact-check: FALSE

Read the linked article. I don’t conclude from that article alone that it’s evidence of anti-conservative bias. I conclude that the hashtag was still on twitter’s flag list, the tweet got flagged, and the account was either auto-banned with no actual review, or some dolt clicked the "ban" button without bothering to look at context.

Could it be bias? Sure. But that one article alone doesn’t convince me.

Having previously worked for a hosting provider and seen content removals firsthand, I feel pretty comfortable stating there’s a WHOLE lot more incompetence going around than bias.

TheLizard (profile) says:

Re: Re: Re:5 Fact-check: FALSE

You’re ignoring the fact that they were that kind of ban-happy over the hashtag in the first place. It was to protect journalists (Twitter’s protected class), who were happy telling laid-off factory workers and miners that THEY should learn to code, but couldn’t take a joke when they were reminded of it after the online rags started mass lay-offs of journalists.

THAT is the bias.

TheLizard (profile) says:

Re: Re: Re:7 Re:

Right, as I pointed out in my 2nd post, a lot of it just has to do with anti-establishment viewpoints.

But, if you’re always protecting journalists this way, you will, almost by definition, have a bias against conservatives, as journalists as a group are generally very liberal.

Comprehensive meta-study on that:


Stephen T. Stone (profile) says:

Re: Re: Re:8

journalists as a group are generally very liberal

“Conservatives as a group are generally very homophobic.” I, too, can make sweeping generalizations about entire segments of the population based on mere anecdotal evidence (e.g., prominent conservative personalities and lawmakers voicing anti-queer beliefs and supporting efforts to marginalize queer people). You know what that makes me? An asshole.

bob says:

Re: Re: Re:4 Fact-check: do it correctly

At least this reply has something to look at for potential evidence. I only say potential because I haven’t read your source material yet. So I can’t definitively state that you have evidence.

The problem before was you just said there’s a problem with X thing. But you didn’t give any details. Well there is a lot of problems with a lot of things so what’s your point.

That is not presenting evidence that’s you wanting others to prove your point without you having to make an effort. You stated an idea with no backing. You must first show evidence and do some research, not someone else.

However in your later reply you actually posted something that at least could be evidence. That is how it works.

Now if what you presented is factually correct then you have a chance that someone will reply. If they just state their opinion refuting you then you can ignore the response because they lack evidence.

But if they reply with their own valid evidence supporting their viewpoint. Now you can debate the merits of each viewpoint and weigh the evidence.

After seeing the evidence hopefully the party that was wrong will acknowledge their error. Then instead of being a dick because their viewpoint was defeated each party can move on.

TheLizard (profile) says:

Re: Re: Re:5 Re:

Right, that’s true.

But there ARE a lot of anecdotes, that do draw a pattern. This was just one of the most obvious and egregious ones.

The Joe Rogan interview with Jack Dorsey, Vijaya Gadde, and Tim Pool is very instructive. Pool points out that their TOS itself is biased against conservatives, which is why one user was permanent-lifetime banned for posting "Men aren’t women, though."


Alice Andalack says:

Re: Re: Re: You can't NOW prove Jews suffered in WW2.

Holocaust Paradox: Long Lives for Those Who Survived…


Goes against all expectation. Despite the alleged horrors, they’re living longer than the contemporary 18 year old soldiers. And so many! Just how effective were the Krauts, anyway?

POINT IS: It’s impossible to prove anything to a biased corporatist. A corporatist which openly advocates corporations having government-conferred power to control even Constitutionally protected speech of course won’t admit that there’s an anti-conservative bias!

You’re trying to shift burden of proof by expanding "rights" of a mere legal fiction. But whatever_corporation, being a mere legal fiction which is let exist in first place on promise to serve The Public, has to prove its innocence, not The Public to prove every detail.

bob says:

Re: Re: Re:2 You can't NOW prove Jews suffered in WW2.

Pulling a Godwin and denying the holocaust happened won’t change or erase what the Nazis did.

Just how effective were the Krauts, anyway?

At least 6 million Jews specifically.

Considering the cruel and abysmal conditions for concentration camps it’s safer to believe that those that survived had a mixture of luck and good genes.

Now go take your obviously trolling behaviour and depart you holocaust denying piece of crap.

Stephen T. Stone (profile) says:

Re: Re:

“Anti-conservative bias” in moderation on social media platforms is a myth. To claim it as a fact, you must prove two things:

  1. Any punishment of a conservative user — and only a conservative user — on a given platform is done only because of their political beliefs.
  2. The platform has a distinct and provable pattern of politically motivated punishments that target conservatives/Republicans/right-wingers but leave liberals/Democrats/left-wingers alone.

I wish you the best of luck; you will certainly need it.

Anonymous Coward says:

Re: Fact-check: FALSE

There are lots of reasons that something might be removed that have nothing to do with bias. I used to work for a company that hosted a website that blamed an earthquake in New Zealand on homosexuals. They got a deluge of complaints demanding the site be removed, but didn’t act based on free speech grounds. It wasn’t until a DMCA notice came in against one of the photos on the site that they did something, which was to completely terminate the account – not specifically for the copyright issue (though that was the excuse given to the customer), but because the company didn’t want to deal with the complaints anymore.

Wendy Cockcroft (profile) says:

Re: Re: Re: Re:

^This. And, as I commented before, "conservatives" have no desire to clean house and sweep out the idiots that make the rest of us look bad. If they kicked out the crazies and refused to let them use the label without being challenged for it, we’d see a difference. The people defending the indefensible need to have a word with themselves.

Alice Andalack says:

Re: BIAS -- @ "TheLizard" -- You're not going to win here...

With mere proof.

Techdirt is proof against proof.

The site is for entertainment purposes only.

I know, you see apparently people who can be reasoned with, but no one has ever changed Masnick’s views. — Took hundreds of thousands of dollars to force Masnick to even put up a link to fair treatment of Ayyadurai’s views! And yet Masnick is back just now with another repeat!

Anyhoo, for your own sanity, don’t take this site seriously.

James Burkhardt (profile) says:


So, this article doesn’t show an anti-conservative bias, a repeated point that has been made. This evidence shows exactly what Mike noted – issues with the vagaries of content moderation at scale. The hashtag was flagged after it became associated in part with harassment campaigns. Not because of a conservative association. This lead to either an automated temp ban or a worker enforcing an inflexible policy (both of which Mike and Techdirt have noted are issues when performing content moderation at scale), not because of the conservative viewpoints but because of the use of the banned hashtag.

Anonymous Coward says:

Spent some years volunteer moderating default subs on reddit.

A volunteer moderator team will ever be overwhelmed with content to review. Burnout necessitates recruiting new volunteer mods. This provides in-roads for conflict of interest. Each new round of volunteer mods brought on included effort to add ambiguous rule-sets to the code of conduct then used to selectively remove perfectly reasonable comments or submissions.

Considerable time spent weeding out bad actors on the mod team. Closed moderator chat conversation became plagued with conflict as bad actors worked to speed up burnout, harass, and try to remove mods that worked to operate with clearly defined rules and observe free speech. By the time I moved on I was convinced digital media figures routinely fight for control of moderator positions in order to promote their own content, and remove competitors from view.

MathFox says:


Also someone with moderation experience. On websites were very few comments were deleted, but you need to do some housekeeping to keep a forum decent. The basis that works:

  • Remove commercial spam
  • Remove comments with foul language
  • Welcome people with differing opinions
  • Feel free to reorganize if a discussion strays too much from the original topic
    Moderation is a tool to keep a forum enjoyable for the visitors… not something to propagate an agenda.
James Burkhardt (profile) says:


Showing Political bias in moderation requires that you show the partisan aspects of the content are the cause of moderation, in addition to showing that moderation happens significantly more often to individuals of a specific partisan affiliation.

The issue with analyzing this is that most social media will provide a biased perspective on who is moderated, due to bias in selecting whom to follow. Those whose views are on the right likely does not follow many whose views are radical and left, and vice versa. That leads to me seeing more complaints about how social media is biased against progressives, seeing mostly the leftists who are unfairly punished and those radical leftists who may clearly fall afoul of TOS, even if I may agree with the views reflected in the posts being taken down. But I mostly am friends with moderate right individuals, so I see far fewer documented cases of moderation of conservatives, with only the ones that make the news cycle really coming up. (though i tend to see a lot of fake "facebook censored this" memes from both sides) This is why a general statistical analysis, rather then personal experience, is important to prove bias.

You talk about wanting facts, but you provide none. Everytime I see ‘proof’ presented to me, I find the proof can be explained by well known issues relating to content moderation at scale that match my own experiences as a forum moderator back in the early 2000’s, without any analysis of if this proof is in fact typical of a platform in general.

ECA (profile) says:


The only real problems with debating or publishing 1 person OPINION, tends to be the idea stuck in a person head..
"Im right you are wrong". attitudes that Any/all sides may have.

biased against progressives
seeing mostly the leftists who are unfairly punished
radical leftists who may clearly fall afoul of TOS
moderate right individuals

Love the wording you are using, But I tend to think of people as individuals first, then ask them where they got there Ideas/ideals.. And most cant answer Where they got those ideas.

Old friend came back after about 5 years wondering the USA. He made a random statement, "OBAMA DID IT". I Asked him "WHAT did Obama Do?? Explain your comment". My friend, Stuttered, abit, so I asked him "Who told you that? since you dont know, what Obama did."

I think there is more Variance in Both sides then we think, esp. with self opinions..I cant see a WHOLE group seeing 1 concept/idea/… in any situation. Dogma be damned, quit listening to what others SAY..Make up your own minds. OR throw your bible away and Just believe what they tell you.

Add Your Comment

Your email address will not be published. Required fields are marked *

Have a Techdirt Account? Sign in now. Want one? Register here

Comment Options:

Make this the or (get credits or sign in to see balance) what's this?

What's this?

Techdirt community members with Techdirt Credits can spotlight a comment as either the "First Word" or "Last Word" on a particular comment thread. Credits can be purchased at the Techdirt Insider Shop »

Follow Techdirt

Techdirt Daily Newsletter

Techdirt Deals
Techdirt Insider Discord
The latest chatter on the Techdirt Insider Discord channel...