As the midterm elections ramp up — and the foreign interference from Russian operatives during the 2016 election continues to haunt pseople — Google, Facebook and Twitter have all announced more stringent rules around online ad disclosures. And the Federal Election Commission too is seeking public input on a proposal that would expand political ad disclosure rules for other media to mobile apps.

So far, though, the disclosure rules at Facebook are not just a work in progress, but more of a train wreck in progress, as flaws with Facebook’s new rules are already emerging. The social giant was supposed to archive all political ads publicly but missed some, and critics have complained that they aren’t giving enough information about how the ads are targeted. And so many publishers are getting caught up in the rules and can’t boost their own political stories on Facebook without jumping through hoops. An effort by the News Media Alliance to get publishers whitelisted has started, but so far Facebook doesn’t have plans to do that.

It’s clear that more overarching, consistent rules across social media and the web — not just from Facebook and other tech giants — are needed to monitor political ads on the internet.

Facebook’s New Rules – And More Hiccups

In May, Facebook launched its ad transparency tool in the U.S., which requires ad purchasers on Facebook and Instagram to reveal and verify their identities and locations. Upon verification, those ads — whether for candidates or political issues — must come with a “paid for by” label disclosing who paid for the ad. Anyone on Facebook can now also search for any of these ads for up to seven years in an online database the company set up.

Almost immediately, the new rules caused hiccups. In California, at least one progressive group was able to continue targeting local voters with negative campaign ads about a candidate without disclosing who was behind the campaign. In Florida, ads for more than a dozen political candidates and organizations — which had cleared previous disclaimer requirements — were shut down because they didn’t reveal who paid for them. This resulted in a loss of thousands of dollars for campaigners already in full election swing mode. Facebook’s new tool also saw an article’s link to a Breitbart piece as a red flag. The article was immediately shut down — even though linking to Breitbart was actually part of a broader initiative to correct partisan propaganda, not spread it.

Meanwhile, the Florida Department of Highway Safety and Motor Vehicles had its safety ads caught up in the terms of the new rules, even though it’s running public service announcements. And certain loopholes allow one person to go through Facebook’s verification process and then give the reins for an account to someone else.

What’s more, Facebook needs to negotiate with unique local transparency laws at the same time it’s trying to implement these rules nationwide and eventually globally. Failure to do so means the social giant is actually violating local laws with these new ad disclosures — which is precisely what’s happening now in Seattle and Washington state, which is now suing Facebook and Google for violating those laws.

Do the Rules Go Far Enough?

While there are obvious problems out of the gate, there are also some important missing elements to Facebook’s rules.

Facebook’s new database is missing ads it should obviously catch, as ProPublica’s Jeremy Merrill, Ariana Tobin and Madeleine Varner have reported. But more strikingly, Facebook is also not revealing a critical piece of information that’s been a contentious force in past advertising: how the ads are being targeted. ProPublica previously found, for example, that certain housing ads on Facebook were made to exclude African Americans and other “multicultural affinities” from seeing them, which violates Fair Housing Act protections.

In response, ProPublica is building an independent database of political ads to monitor Facebook’s efforts.

Publishers Struggle

While certain roadblocks are to be expected in the process of setting things up, publishers have to be especially alert to Facebook’s new rules. The definition of “ads with political content” is “astonishingly broad and may affect any promoted content that relates even vaguely to the political system,”  said Ned Berke, program director of audience development at the Center for Cooperative Media at Montclair State University.

Here are some particular points publishers should keep in mind to avoid scuffles, according to Berke:

• get authorized;
• bookmark the 20 issues of public importance Facebook has outlined;
• modify ads that started running before May 24 — before Facebook’s new terms launched — so they have the required disclaimers;
• watch out for the keywords that could trigger Facebook’s machine-learning classifier.

Many angry smaller publishers have decided not to boost posts or run Facebook ads until the social giant makes it easier for them to get verified or starts a whitelisting regime so they can promote their journalism covering politics.

Self-Regulation?

While the FEC is currently waiting on public comment for its proposal of new disclaimer rules for online political ads, the Digital Advertising Alliance is advocating for more self-regulation that could potentially work in conjunction with the FEC’s proposal. The DAA hopes to create new “Political Ad” icons that would link back to sites disclosing the identity and contact information of advertisers. It would be comparable to the DAA’s AdChoices, an icon that informs consumers of tracking data and gives them the option of opting out of behaviorally targeted ads.

Like the FEC’s proposal, it’s unlikely the DAA’s suggested political ad icon could come out in time for this year’s midterm elections. But the broader point is that we need more compliances on a federal and organizational level aside from just companies making their own rules for their platforms. The Honest Ads Act, which would bring online advertising disclosures on par with television ads, is another push in this direction, though the bill hasn’t gone far in the current Congress.

The Good, The Bad, The Hopeful

Facebook knows it may not catch all ads and has welcomed feedback on actors that appear to be trying to game the system. It’s also stated that political ads will be subject to human review on top of the machine-learning classifiers. The company is also in talks with groups, including news publishers, about how the authorizing, labeling and archiving of political content will affect them. However, the specifics for how publishers can avoid being bogged down in disclosures that may affect their online presence have yet to be announced.

Still, the ad tool and the database are currently confined to the U.S. And in countries like Sri Lanka, Facebook has already proved to be the gasoline that ignites political wars on social media. While the company maintains it’s committed to expanding these initiatives internationally, there’s still a long way to go. While the hope is that things will improve over time, Facebook needs to tread with care and get this balancing act right to make sure trusted publishers can promote their important political coverage. Let’s also hope oversight from independent groups and government regulators comes through as well.