Information split
Illustration: Christopher T. Fong/Protocol

Tech’s not ready for the post-Roe infowar

Protocol Policy

Hello, and welcome to Protocol Policy! Today, we’re talking about tech platforms’ responsibility in fighting misinformation in a post-Roe world. Plus, the Biden administration strikes a deal with ISPs, and Trump loses a court battle.

Post-Roe reality

It took about a day from the leak of the Supreme Court’s Roe decision for internet sleuths to decide they’d figured out whodunnit.

It started with a Twitter thread, which sent searches for the clerk’s name soaring moments later. Pretty soon, posts about the alleged leaker were lighting up Facebook. According to social media monitoring firm Brandwatch, mentions of the alleged leaker’s name (which we’re not sharing here) spiked by 7,300% within a few days.

The Supreme Court marshal is now investigating the leak, and it’s anyone’s guess whether this social media-fueled theory will turn out to be true. The question is: What responsibility do platforms have in the event that it’s not?

The Roe decision is about to become tech’s next big infowar — and charting a path forward isn’t going to be easy. Take the leaker posts.

  • In some ways, they’re similar to the virtual mobs that have descended on alleged terrorism or criminal suspects in the past, who turned out not to be suspects at all.
  • After the Las Vegas shooting in 2017, both Facebook and Google helped boost a bogus 4chan theory that named the wrong man as the shooter. In that case though, both companies were able to make quick fixes once the real shooter’s identity was revealed.
  • This time, there’s no telling if or when the identity of the SCOTUS leaker will actually be known, giving the platforms no firm basis on which to act.

It’s not just about the leaker. It’s also unclear where platforms will draw the line when it comes to medical misinformation about abortion. It’s already popping up in ads.

  • The day after the Roe decision leaked, the anti-abortion group Live Action began running a Facebook ad warning “Abortion pills are not safe.” This is despite the FDA’s assurances that they are. Does such a claim violate Facebook’s rules against misinformation in ads? We asked. Meta wouldn’t say.
  • The same questions apply to supposed “abortion pill reversal” services, which the American College of Obstetricians and Gynecologists says are “not supported by science.” Would ads for those services be barred as “speculative and experimental medical treatments” by Google? Google also wouldn’t answer our questions.

It’s no big surprise that tech giants aren’t interested in drawing bright lines around one of the most polarizing issues in America. Meta, Twitter and Google all directed Protocol to their existing policies — but as in all things, your interpretation of how those policies apply to abortion misinformation probably depends a lot on your views about abortion.

None of these questions are exactly new. Pro- and anti-abortion groups have been fighting it out on social media for years, and fighting it out with picket signs for even longer. But with in-person clinics expected to evaporate across wide swaths of America, the ability for people to get reliable information about abortion online has arguably never been more critical.

— Issie Lapowsky (email | twitter)

In Washington

Twenty ISPs will dramatically cut costs for low-income Americans after striking a deal with the Biden administration. The providers, including AT&T, Comcast and Verizon, will offer high-speed plans for no more than $30 a month. That deal, combined with the $30 monthly broadband benefits in the infrastructure law, means eligible Americans should be able to get high-speed service for free.

In the courts

A federal judge dismissed Trump’s “censorship” lawsuit against Twitter, which argued that the company had violated the former president’s First Amendment rights. “Plaintiffs are not starting from a position of strength,” the judge wrote.

Tesla is suing a former engineer, alleging he stole confidential information about Tesla’s efforts to build an in-house supercomputer. The company says the engineer then tried to pass off a “dummy” computer for inspection.

Oral arguments began in Texas’ appeal of a ruling against its social media law. The law was already blocked on the grounds that it was unconstitutional after Texas was sued by NetChoice and the Computer and Communications Industry Association.


New polling shows that American voters do not see regulating tech companies as a priority. Their top concerns are strengthening the national economy (38%), followed by controlling inflation (37%). By contrast, only 5% of respondents prioritized regulating tech companies.

Learn more

On Protocol

The fossil fuel industry is pushing a new (factually dubious) narrative on Facebook. According to an analysis by Media Matters, some of the industry’s most popular recent posts focus on the need for “energy independence” to justify more drilling, even though the U.S. is a net exporter of liquid natural gas. Only two of the 100 posts analyzed had climate misinfo labels attached.

Could labeling junk news discourage people from reading it?Not likely, according to new research from NYU. The study found that rating news outlets by reliability didn’t nudge most people to more reliable outlets — unless they were among the most voracious readers of unreliable news to begin with.

Around the world

Google is blocking paid app downloads in Russia, citing “payment system disruption.” While sanctions prohibit certain financial transactions, free apps and other free Google services are still available there.

COVID-19 restrictions in China have forced Apple to rely on local engineers to oversee its manufacturing operations in the country. The company had previously sent U.S. engineers to China to do the work, but limits on foreign visas during the pandemic prohibit that kind of travel.

In the media, culture and metaverse

An analysis of Russian state media shows how the country is using its vast propaganda network not only to mislead Russian people about the war in Ukraine, but also to confuse them to the point of not knowing what to trust.

Former U.S. TikTok employees say they averaged 85 hours of meetings a week while working for the company, which has adopted a grueling work environment, according to the Wall Street Journal. “If I knew that working at TikTok would cost me this much, I would never have taken the job,” one former employee wrote in a Medium post.

In the C-suite

Uber is cutting costs and treating hiring as a “privilege,” according to a leaked email to employees from CEO Dara Khosrowshahi. Khosrowshahi wrote that the changes are a response to “a seismic shift” in the market. Meta made a similar declaration about hiring last week.


New polling shows voters' top tech policy concerns are cybersecurity and data privacy. Only 7% of respondents prioritized antitrust action and 1% prioritized changes to app store rules. In fact, the majority (58%) believe the pending tech antitrust legislation would cause more harm than help to consumers.

Learn more

Ban the bot that feeds you

Elon Musk might want to think twice about banning bots on Twitter: They’ve been awfully good to Tesla, after all. According to The Washington Post, researchers at the University of Maryland found one account in particular, @danrocks4, that has posted about Tesla, on average, every three hours for more than six years.

Thanks for reading — see you Wednesday!

Recent Issues