Twitter's Decision To Ban Political Ads Is A Moderation Choice Itself That Likely Will Backfire In Its Own Way
from the because-content-moderation-is-impossible-to-do-well dept
Last week we wrote about Twitter's decision to ban all political ads, most likely in response to watching all the shit being flung at Facebook for its decision to not fact check political ads. We focused on the fact that the "costs" of content moderation can sometimes be so high as to make any related revenue just not worth it. However, in that post we did mention that no matter what, there would be criticism of this decision and follow-on decisions concerning what is, and what is not, a "political" advertisement.
There have been a bunch of good, thoughtful articles about all of this that seem worth highlighting. First up is a piece from Markena Kelly at the Verge, who pointed out that Facebook has already tried to ban political ads, but just in the state of Washington, in response to local laws. And just as we predicted will happen with Twitter, there have been ongoing disputes over what constitutes a political ad:
The first major test case for the new system came with Seattle’s city council elections, which will be wrapping in November. Marijuana entrepreneur Logan Bowers ran for city council on an urbanist platform, but he ended up fighting an uphill battle on platforms. He says confusion around the ban “created an unfair and an unlevel playing field and in many ways it made the situation worse.” High-profile ads were ultimately removed by Facebook, usually after they were reported in the media — but plenty of others skated through.
“Some people had their ads restricted and other people didn’t,” Bowers says, usually according to who knew how to spot the loopholes in the system. “Not everyone’s a lawyer.”
And, as that article notes, even though Facebook did this to try to comply with local state laws, the state is still going after Facebook. So it didn't even help on that front. Instead, it was just fights over whose ads get through:
In April, The Stranger reported that one Seattle City Council candidate, Heidi Wills, was able to run a handful of ads on Facebook while her opponent, Kate Martin, was blocked from running any. The two candidates got into a spat through the Wills campaign’s own comments section on Facebook with Martin pleading, “Could you stop paying to promote your Facebook posts and just play by the rules like the rest of us? It’s getting annoying.”
Wills replied, “I am following all the rules and you are welcome to stop following my campaign on FB.”
Meanwhile, Will Oremus, writing for OneZero points out that Twitter's ban is likely to mostly harm activists and organizers:
There’s something to be said for a tech platform taking its responsibilities to the democratic process seriously. But banning political ads is not as straightforward, nor as obviously correct, as those cheering Dorsey’s announcement seem to think.
The problem is twofold. First, defining which ads count as “political” gets tricky in a hurry. Second, prioritizing commercial speech over political speech is itself a political stance, and not necessarily one that we should want our online communication platforms to take.
He notes all sorts of potential downstream problems:
Presumably, tech companies will still be able to run ads touting their commitment to user privacy, but watchdog groups will be barred from running ads suggesting that we need better privacy regulations. Big corporations will be able to boast about how they treat workers, but unions won’t be able to push for prevailing wage laws or workplace safety laws.
Meanwhile, Cat Zakrzewski, writing over at the Washington Post, speaks to some experts who also note that figuring out what counts as a political ad is very, very difficult, leading to plenty of gaming the system (as Facebook discovered above):
Laura Edelson, a PhD candidate at New York University studying political ads on social media, tells me she found instances in the 2018 election where ads from multiple sitting senators were not marked as political ads by Twitter.
That report notes the general problems with classifying ads at all. And while it's framed as something that Twitter is bad at, the reality is that anyone is likely to be bad at this, since so much of this relies on subjective calls, and no one's going to agree.
“The technical problem of enforcing this ban is the same one as enforcing their disclosure requirements, and if they make less information transparent, it will be harder for third parties like us to monitor if they are actually enforcing this policy,” she added.
Shannon McGregor, an assistant professor at The University of Utah, said she and other researchers detected instances where foreign governments were running ads on Twitter subject to the Foreign Agents Registration Act — yet those ads were not identified in Twitter’s ad database. She said if Twitter was unable to identify and include ads from a foreign government in its library, it raises questions about the site's ability to ensure that no political ads about domestic issues are run on its platform.
Meanwhile, the chair of the Federal Election Commission, who in theory has some purview over political advertising, has weighed in as well with a pretty smart take about how banning all political ads doesn't really make much sense. Instead, it might make much more sense to limit the "microtargeting" of such ads.
Here’s a move that would allow political ads while deterring disinformation campaigns, restoring transparency and protecting the robust marketplace of ideas: Sell political ads, but stop the practice of microtargeting those ads.
“Microtargeting” is the sales practice of limiting the scope of an ad’s distribution to precise sets of people, such as single men between 25 and 35 who live in apartments and “like” the Washington Nationals. But just because microtargeted ads can be a good way to sell deodorant does not make them a safe way to sell candidates. It is easy to single out susceptible groups and direct political misinformation to them with little accountability, because the public at large never sees the ad.
As she notes, this approach would enhance transparency and accountability (since people could better see what political ads were on the platform), enable the ability for more people to call out and debunk disinformation, and (in theory -- perhaps optimistically) push political advertisers to create messages that apply to a broader group of people, rather than narrowly targeting certain groups and "fueling the divisiveness that pulls us apart."
Of course, to some extent this still relies on Twitter being able to decide what is and what is not a political ad. We've argued right here in the past that social media platforms should consider dumping targeted ads altogether, but people always yell at us that such ads are way too valuable to completely dump. Still, bringing things back around to our first post on Twitter's decision: perhaps if the "costs" of continuing to run such ads is so high, they'll find that it makes more sense to dump them entirely.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: content moderation, fact checking, political ads, scale
Companies: twitter
Reader Comments
Subscribe: RSS
View by: Time | Thread
Simple Fix!
All twitter needs to do is put a little checkbox on the form you fill out to purchase an advertisement.
"Is this a political ad?"
If the box is checked, you can't proceed with purchasing the advertisement. There you go! Everything has a simple solution if you just nerd harder.
[ link to this | view in chronology ]
Re: Simple Fix!
You're always so roundabout, Gary.
Just have Dumbledore or Gandalf programming the thing to determine whether the ad's political or not. Problem Solved!
[ link to this | view in chronology ]
Small consolations...
One thing that the proliferation of conspiracy theories, fake news and astro-turfed moral panic shows is that false advertising has become just as democratized as other forms of media on the internet.
Now anyone can push their special interest agenda on the internet and have a decent chance of having it spread. False advertising is no longer just the realm of ISPs, nutritional supplement vendors and political parties. With the right tools and a small investment you too can project your delusions to hundreds, thousands or possibly millions of people. This is much more access than anyone but the most influential people had in the days of cable TV or the big 3 networks.
[ link to this | view in chronology ]
Re: Small consolations...
Don't know that I would call it "democratized".
More like the marketplace has been monetized.
[ link to this | view in chronology ]
A Simple question and response to solve the problem
Are you doing this in ways that could Be construed as political?
Awesome GTFO no one wants your kind here! All you do is shatter dreams of children and innocents! You took the internets promise and smashed it again-“continues for some time”
[ link to this | view in chronology ]
A choice to not moderate at ll is also a choice of moderation. Twitter;s choice in this case is at least an attempt to minimize abuses. It may not be ideal and it may have some unintended consequences, but developing a tractable solution that achieves some desired results, like truth in advertising, is going to be an iterative process. No solution to anything comes about on the first shot.
To the extent one invokes freedom of speach here, recall that fraudulent advertising for non-existent products, lying on IRS forms, etc., are also speech - just speech that seemed reasonable to create statutes to regulate. What I think no one wants are statutes to regulate online speech, which is what is likely to happen if companies like Twitter don't make some attempt to create a tractable scheme to reign in some of the abuses (which as a private company is certainly within their rights).
What I do not want is the technically illiterate idiots in congress deciding to write laws that create court battles that drag on for decades, which in many cases seems to end up with the government winning. If there are issues with what Twitter is doing, suggest ways to improve it. the only way to make the GDPR worse would be to have the US congress reimplement it here.
[ link to this | view in chronology ]
Re:
But this choice doesn’t minimize abuse, that’s the problem.
You may think that this specific set of abuse is better than another, but it’s not a minimum.
To see it, you have two choices. Two guys are walking up to you, if you pick Option A one of them will always punch you in the face (and the other may, or may not), if you choose Option B each of them might (and might not) punch your in the face. Choosing Option A always leads to more abuse because it is itself abusive.
The same with this: no politics. So you can’t run an add for your GoFundMe to pay for a lawyer to help get you out of a CIA torture center because that’s petitioning the government, and governmental action is inherently political. This is, of course, a really edge case (I hope!), but it illustrates that a no-politics rule will always harm the person who wants whatever the result of politics are. You could have said the same in 1964 - no ads arguing for why we needed the Civil Rights Act.
All that this sort of rule does is shift who is harmed, how they’re harmed, and how clearly we can see the harm. If you think you hold an objectively superior view then you should want all political ads because objective truth will overcome sophistry, unless you don’t believe in democracy (which is at least internally consistent, of course - oligopolies have a lot in their favor, if you’re an oligopolist).
[ link to this | view in chronology ]
Re: Re:
And if I ran Twitter, I would personally opt for eliminating incoherent arguments which offer no logical structure nor attempt to demonstrate any basis for whatever is being claimed.
[ link to this | view in chronology ]
The whole world should ban political ads. Ifor one am sick of all their lies and deciding they are above the law and giving themselves outrageous raises whenever they want. I would love for the entire world to stop giving them a voice period.
[ link to this | view in chronology ]
Re:
So you are saying that it's time to campaign for a new law to ban them.
Lets start a campaign to pas this!
[ link to this | view in chronology ]
Re:
As long as that also includes those fugly political signs slapped all over roadways and yards, I'm totally on board.
[ link to this | view in chronology ]
Re:
Take all the money spent on Political adds and put it into something useful, like providing universal Healthcare (we could provide 5* healthcare to everyone in the world for the cost of all the political adds).
If you don't support this, then you are obviously part of the problem, not the solution...
Lets also Bayh-Doyle the shit out drugs and make all the generic versions freely available, with no 'extended patent life' payments to the middlemen (most drugs are created by public funds, so the middlemen are just getting rich off of others work anyway).
There we go, universal healthcare for all and freely available drugs... solved the world's problems before lunch even...
[ link to this | view in chronology ]
Re:
I don't know where you live but, in the United States, no pay change for Congress can go into effect until after the next Congressional term begins. That's laid out in the Twenty-Seventh Amendment.
By law, very few people employed by the US Government earn more than the President does, and his (or her) salary creates the scale for most employees. The few employees who earn more are those in fields where the average salary is very high already (coaches at the Division I service academies and some surgeons in the VA). So, if we want the lower-level employees to earn enough to make government service appealing compared to private-sector employment, the pay at the top end has to go up too.
Lastly, the US federal government has one of the better "CEO" to bottom level employee pay ratio's in the country. Since the President is paid $400,000/year and the minimum hourly wage for a federal employee is $10.35/hour, assuming full-time employment for that bottom end employee, the ratio between them is only about 18.6. When you look at the wages of some CEO's compared to their least paid employees, it's easy to see those ratios get into the hundreds.
[ link to this | view in chronology ]
Thank you R.H !!
I had never learned of the amazing background to the 27th amendment. Its a ripper of a story:
https://en.wikipedia.org/wiki/Twenty-seventh_Amendment_to_the_United_States_Constitution
[ link to this | view in chronology ]
Re: Re:
Congress gets bad press every time they give themselves a raise, and yes - poor babies, they are forced to wait for it.
Ralph Nader once pointed out to congress that their proposed raise was more than the mean income of the country's population. They passed it anyway. One dude said it was difficult owning two houses and commuting back n forth ... as if he was unaware of that when he ran for office.
[ link to this | view in chronology ]
Banning Political Ads is a Moderation Choice...
...and it works at scale. Still waiting for my apology Mike.
[ link to this | view in chronology ]
Re: Banning Political Ads is a Moderation Choice...
There is a very fine line between smartass and dumbass.
[ link to this | view in chronology ]
Re: Re: Banning Political Ads is a Moderation Choice...
...and either way, you're still an ass.
[ link to this | view in chronology ]
Re: Banning Political Ads is a Moderation Choice...
"Banning Political Ads is a Moderation Choice...
...and it works at scale. Still waiting for my apology Mike."
Hahaha, yeah.
However - upon futher thought, how would it work at scale when no one can tell what a political ad is?
[ link to this | view in chronology ]
Re: Banning Political Ads is a Moderation Choice...
[Asserts facts not in evidence]
Seriously, would it kill you to actually read an article before trolling it?
[ link to this | view in chronology ]
The initial annoucement said ads about specific issues, not just political ads, would be blocked. Of course, that just pushes the problem elsewhere. In my choice of soft drink an "issue"? Hell, it's explicitly political in some areas (various US cities have passed sugary drink taxes). Gas-guzzling vehicles and electric vehicles can't be disentangled from their climate-change implications. TV shows have political subtexts. What could you advertise?
[ link to this | view in chronology ]
Re:
well the people who pushing the waters these ways don’t really care so long as they get what they want.
“Looking at you warren you fake native”
[ link to this | view in chronology ]
Fake nativity
I'd really like a political climate where dubious heritage and consensual adult relationships (whether with porn stars or interns) mattered less than histories of sexual assault, torture policies, massacre policies, extrajudicial detention policies, the nepotism / incompetence loop, secret laws, secret courts, secret mass surveillance programs...you probably get the idea.
It pains me that all the shots taken, at candidates, at officials, even at derided minority groups, are cheap.
[ link to this | view in chronology ]
Re: Fake nativity
It’s a sad commentary of what people think works that the cheap shots are taken when (essentially) anyone running for public office has literal skeletons in their metaphorical closet.
Trump, at least, is brazen about his vulgarities. It’s like when a bartender tells you he swapped out the top shelf for grain alcohol because at this point you can’t tell the difference. Is it more honest to be told to your face you’re being screwed, or to be screwed behind your back?
[ link to this | view in chronology ]
What!?
Anyone running for public office has literal skeletons in their metaphorical closet.
When I read that it suggest all candidates are guilty of murder (or complicit to manslaughter?) But that blood is on their hands.
Trump, at least, is brazen about his vulgarities...
Trump is brazen about his vulgarities because he is psychologically impaired from behaving with decorum. He's also psychologically impaired from being able to adhere to the truth during a deposition, which is the reason Mueller couldn't get an oral testimony, and had to rely instead on a written one prepared by Trump's lawyers. They might have been able to make a diminished capacity appeal except that doing so would admit his unfitness to hold office.
Trump is also grotesquely dishonest about facts and his intents, having recently exceeded 13,000 statements through his presidency so far.
Trump in office may be a circumstance that serves you personally, but it's also an indictment of the US election system, having demonstrated that all the mechanisms intended to keep puppets, madmen and tyrants out of office have failed and need to be reconsidered.
[ link to this | view in chronology ]
Re: Re: Fake nativity
The power of negative thinking
[ link to this | view in chronology ]
Re: Re:
Of all the things you could complain about Warren over why choose the racist one?
[ link to this | view in chronology ]
Re: Re: Re:
I think it may be all they've got.
[ link to this | view in chronology ]
The least bad easy solution.
Banning political ads (political ads I assume includes any paid advertisement that clearly advocates for or against a candidate or policy) seems like an easy action Twitter can take that has fewer repercussions (for Twitter) than not doing it.
It's not going to stop political ads (An banner advertising an alleged sale of AR-15 style rifles -- EVERYTHING MUST GO! -- taking place on an elementary school playground would be politically inciteful while looking like it's, well, selling munitions. We might see some very clever non-political, political ads (ALL ABORTIONS FREE THIS WEEKEND AT OUR CLINIC!), and I will personally be amused by art inspired by creative constraint.
It also doesn't stop someone from tweeting their opinion-shaped-advert and then hiring a Chinese zombie bank to retweet it a gajillion times, to then let the Twitter magic take care of the rest. I suspect my simplistic understanding of Twitter magic (or websearch magic) is limited, but the process is still there.
The problem remains that voters are not the rational self-interested thinkers that Jefferson imagined them to be, but easily manipulable in numerous ways, and with free speech, society is exposed to that manipulation used to seize control (or cause discontent and chaos).
But the same question that comes with munitions, political franchise and religion comes with political speech: if none of us can be trust to be adults in the room, who do we trust to be the adult in the room? Who do we trust with power we cannot trust to ourselves or each other?
It's not a question for which I've found an adequate answer. I like my bread to be free of sawdust and microplastics, but I also don't trust the FDA to adequately keep it out of my bread.
[ link to this | view in chronology ]
Re: The least bad easy solution.
No matter what you do, some fraction of people will not be satisfied. So, if, for example, what Twitter is doing is problematic for enough people, there is always room for another forum to appeal to those people. If no one does anything, the most likely outcome will be the government stepping in to create the worst of all possible worlds.
If the question is about what is best for society, then different people will have somewhat different opinions about what that is, but even the best result is that most people will agree that even if they didn't get everything everyone wanted, that overall it fits the ideas about society embedded in the constitution.
[ link to this | view in chronology ]
Re: Re: The least bad easy solution.
Really there are two problems.
[ link to this | view in chronology ]
Re: Re: Re: The least bad easy solution.
Also, if politicians want to place ads without restrictions, they should try 4chan or 8chan. For the majority of candidates, their ads would be pretty much on the same level with the rest of the posts on those sites.
[ link to this | view in chronology ]
Re: The least bad easy solution.
The solution to the bread problem at least is easy - liability, including criminal liability.
And if you think that some brand will just run up their reliability and then skimp out and start cramming in junk, then add a qui tam law like the False Claims Act has that allows private prosecutions by company employees. You work for a bakery that said they don’t use sawdust, but you have a memo saying to use it? Sue as a qui tam relator and take them to the bank.
The problem is where people fundamentally disagree, not on the desired outcomes, but on the predicates. Gun control is a good example: pro gun rights people think it protects the on the small scale (which is at least disprovable if false) but also on the macro scale - no government has ever massacred its population before seizing guns, so its a canary in a coal mine - if you still have strong gun rights you don’t need to be too worried about government tyranny. Anti gun proponents want fewer gun deaths (just like gun rights advocates do) but think guns won’t help the individual and that the government won’t massacre the people. When the argument stays revolving around “you have blood on your hands” no one can ever be persuaded because the argument being made isn’t one that could ever work - they’re not in disagreement on saving lives, they’re in disagreement on what methods will save lives - and in the gun example is unsolvable because we don’t know what has to be done to stop governmental mass murders.
[ link to this | view in chronology ]
Re: Re: The least bad easy solution.
Um, food companies use the same strategies as the military and government departments do to quash whistleblowers, and we have three digits of whistleblowers in jail doing sentences that make murderers blush. Essentially, if I filed a consumer protection warning regarding a Coca-Cola product, the Coca-Cola company would make sure I lost everything and never worked again. And to Hell with what laws say they're allowed to do, as the legislators, regulators and courts are already on their payroll.
The thing is, only in maybe the last twenty years has our bread been consistently without sawdust, not that we knew that our bread had sawdust in it until, well... the information age. In the meantime, there's a pharmacopia of food additives they're trying (fillers, preservatives, agents to perfect the mouth-feel, whatever) and unless people drop dead right away (or are pooping their pants), we won't know.
Also the FDA adjusts its speed in approving / obstructing products based on the policies of the current administration, which is why countless European birth control and feminine hygiene options are banned in the US, and we've had a long line since the 1990s.
Regulatory capture is a government failure. And that is a thickener holding us fast in the quagmire we're in today.
[ link to this | view in chronology ]
Re: Re: The least bad easy solution.
Well, in the case of gun control, (and political advertising as well) it's more politically expedient to argue about the symptom than the disease.
[ link to this | view in chronology ]
Damned if you do, damned if you dont, Damned to even try.
[ link to this | view in chronology ]
Re: Damned if you do, damned if you dont, Damned to even try.
ECA you still trying to slot a nap into your schedule?
[ link to this | view in chronology ]
Personally, I think a possible idea is that political ads run by the politician or their official campaign would be banned, while political ads from anyone else would be treated like any non-political ad and fact-checked accordingly. Not everyone would be happy, but it would avoid the mess of trying to define a political ad.
[ link to this | view in chronology ]
But Super-PACs
These days major federal elections are run less from the candidate's campaign fund rather from Super-PACs which, by definition, are not directly associated with a politician (even when it's super obvious to everyone they totally are).
Is it (by our assessment) on brand for President Donald Trump to use an allegedly unaffiliated Super-PAC to blitz Twitter with ads?
(Donald Trump will from now on be America's hypothetical penetration tester for social and political algorithms.)
[ link to this | view in chronology ]
Re: But Super-PACs
Like I said, those should be fact-checked like any other ad.
[ link to this | view in chronology ]