###

When Sophie Szew first downloaded Instagram at her 10th birthday party, she was exposed to a flurry of information that “promoted eating disorders,” she told California lawmakers. By 15, she said, she was following “every starvation regimen recommended” by Instagram’s “explore” page.

Szew, now 20, spoke in Sacramento at a Senate hearing in April in support of an expansive bill making its way through the Legislature. It would hold companies legally responsible for using algorithms and design features that addict young people.

“Standing with me today is a generation that knows all too well what it is like to be harmed by flawed systems,” she said.

The past few years have seen a steady drip of research and reporting about the effects of social media on teens. That has translated into a stream of legislative activity across the country, with several states passing or considering laws that regulate how social media companies do business.

In 2022, when researchers for a nonprofit created TikTok profiles posing as 13-year old girls in the U.S., U.K., Australia and Canada, they scrolled through videos served to them, stopping to watch and like videos about eating disorders, body image, and mental health. They found that TikTok recommended suicide content within 2.6 minutes and eating disorder content within 8 minutes.

In 2021, after a whistleblower leaked documents, the Wall Street Journal reported that researchers at Meta, formerly known as Facebook, found that 32% of teen girls said that when they felt bad about their bodies, Instagram made them feel worse. Meta owns Instagram..

CalMatters reached out to Twitter, Reddit, Mastodon, TikTok and Meta, which also owns WhatsApp, for comment. Twitter’s press inbox responded automatically with a poop emoji. Reddit, Mastodon, and TikTok did not respond.

In a statement, Meta’s Global Head of Safety, Antigone Davis, wrote that Meta wants teens to be safe online: “We don’t allow content that promotes suicide, self-harm or eating disorders, and of the content we remove or take action on, we identify over 99% of it before it’s reported to us.”

Rachel Holland, a Meta spokesperson, offered information on a variety of features teens and parents can use to shape what young people see, like a setting that makes new teen users less likely to encounter sensitive content, and a feature that nudges teens when they’ve been scrolling through the same kind of content for a while.

What the California bill does

The bill takes aim at how social media companies serve up content to young people under 18. Specifically, it prohibits social media companies from using a design, algorithm, feature, or practice, that companies know (or should know) causes young people to:

• Develop an eating disorder, inflict harm on themselves or others, or become addicted to the social media platform;
• Receive content that facilitates the purchase of controlled substances, like opioids; that facilitates suicide by offering information on how to die by suicide; or facilitates the sale of guns illegally.

It covers social media companies that earn more than $100 million in revenue per year and have users in California. Companies that violate the law could get sued by public attorneys, and would face penalties of as much as $250,000 per violation.

But the bill also builds in a way for companies to protect themselves from lawsuits: By auditing their designs, algorithms, features, and more at least quarterly for their potential to cause the harms listed in the bill and correcting issues within 30 days.

“If you look at a bill like this, the crux of the problem is content.”

— Sophie Cope, Senior Staff Attorney, Electronic Frontier Foundation

California isn’t the only state where legislators are aiming to regulate social media platforms. In March, Utah’s governor signed a similar bill into law, plus another law that institutes a social media curfew for Utah teens under 18 and requires teens to get their parents’ permission to set up social media accounts. Arkansas also passed a law requiring parental permission for teens to set up social media accounts. State legislators in New Jersey are considering a bill that prohibits social media companies from using features that addict kids, and lawmakers in Minnesota are also mulling a social media bill. Then, there are the bills that would regulate social media with a different aim: Texas passed a law that made it illegal to ban users based on their “viewpoints,” and Florida also passed a law restricting platforms’ ability to ban users.

On top of all the new proposals, there are a bevy of lawsuits ping-ponging their way through the courts that could reshape social media regulation. The Texas and Florida laws were challenged for violating the First Amendment; the Supreme Court will decide their fate. More than 80 cases have been grouped together in one jumbo lawsuit, in which plaintiffs say that the social media platforms are essentially defective products — like an exploding toaster — because they addict children.

California bill could face legal hurdles

If the California bill became law, would it be able to dodge and weave its way through legal challenges?

Tech industry lobbyists, as well as some digital rights advocates and internet law experts, say that the bill would immediately bump up against the First Amendment of the U.S. Constitution and federal law. Both outrank or “pre-empt” in lawyer-speak, California laws.

The bill runs afoul of the Constitution because the Supreme Court has generally interpreted the First Amendment as protecting the editorial decisions of publishers, said Sophie Cope, a senior staff attorney at the nonprofit Electronic Frontier Foundation, which opposes the bill. Newspapers decide their layout, what goes on the front page and which images to use, Cope said. Social media platforms are doing similar things as online publishers when they make decisions about how content is displayed, shared, or promoted, she said.

This bill “goes directly at the heart of the editorial discretion of these platforms,” Cope said.

“It’s the First Amendment that restricts the ability of legislators to control the flow of information at children” to the degree legislators might want, agreed Eric Goldman, a professor at Santa Clara University School of Law and an expert on internet law. His example: a Supreme Court case from 1997 about whether a law prohibiting the transmission of porn or “obscene or indecent” messages to kids violated the first amendment. The Supreme Court decided unanimously that it did.

Ed Howard, senior counsel for the Children’s Advocacy Institute at the University of San Diego School of Law and a supporter of the bill, disagrees, and argues that the bill doesn’t violate the First Amendment, in part because the algorithms that prioritize what content gets displayed don’t have their own First Amendment rights. And, he said, “there is no blanket immunity to harm people through your speech.”

Then, there’s a federal law that could pose a problem for the California bill. Section 230 of the Communications Decency Act generally protects tech companies from legal responsibility for content posted by users on their platforms. If, for example, someone posts a defamatory review of a restaurant on Yelp, the restaurant owner could sue the poster, but Yelp itself would be protected, said Cope.

So, a key question is whether the California bill is taking aim at features and algorithms that a social media company has designed — for which it may not be protected by Section 230 — or if it’s really requiring platforms to be accountable for content that users upload.

State Sen. Nancy Skinner, an Oakland Democrat who authored the bill, argues that the bill holds companies liable for their own products and practices, so it wouldn’t be preempted by Section 230. Part of the government’s job, she said, “is to protect consumers from harmful business practices. That’s a very classic role of government.”

But, said Cope with the Electronic Frontier Foundation, “if you look at a bill like this, the crux of the problem is content.” The bill is structured to make platforms “responsible for bad content reaching young people,” so the federal law should pre-empt it, she said. The features, in other words, aren’t a problem when they’re showing #vanlife and knitting videos to young users, but when they’re serving up starvation diet tips, they are.

Public may be left in suspense

Last week the bill was sent to the Senate Appropriations Committee and added to a special pile of bills that the committee will analyze for costs and benefits. That pile of bills — known as the “suspense file” — is officially about budget analysis. Unofficially, it’s also a politically expedient place for lawmakers to kill bills without taking any heat. That’s because, unlike other votes that state legislators take, votes on suspense file bills are kept secret. If a bill never makes it out of the special pile, the public doesn’t know whose votes led to its demise.

A version of the bill was introduced last year. No legislator voted against it publicly, but the bill was put on suspense file and died there.

This year’s bill “could suffer that fate,” said Skinner, “but I’m hopeful that the amount of increased awareness, increased evidence and action by other states will hopefully put it in much better shape,” she said.

###

CalMatters.org is a nonprofit, nonpartisan media venture explaining California policies and politics.