New York adds speed bumps to social media with tobacco-style warning labels, inviting litigation

After years of struggling in federal court to defend its regulation of speech online and in the real world, New York is getting creative in how it seeks to dissuade purportedly harmful words and behaviors, particularly through obligations on tech platforms.

The Empire State's next gambit is treating social media itself as an "addictive" substance, and its core components as addictive features, to coerce platforms into speech that denounces their own products and services when served to users in New York.

Democratic Gov. Kathy Hochul set the stage for a litigious 2026 by signing a bill into law (S4505/A5346) that treats social media like tobacco and alcohol, requiring unavoidable warning labels on "predatory features such as algorithmic feeds, endless scroll, autoplay, notifications, and 'likes'" that must be shown prominently whenever users visit platforms.

As with the recent flood of social media regulation from the U.K. to Australia, Hochul justified the law as "protecting our kids," pointing to former Surgeon General Vivek Murthy's report on social media and youth mental health and its claim that research shows double the risk of anxiety and depression in adolescents who spend more than three hours a day on social media.

But her office also portrayed the law as simply mandating one form of "consumer warning labels" that are routine on even mundane products and services, such as child-suffocation warnings on plastic packaging and flashing-lights warnings on video games that could trigger "users with photosensitive epilepsy."

Hochul deemed herself instrumental to the final language of the bill, saying she "negotiated a chapter amendment" that lays out how the system works. It gives sweeping power to the attorney general to decide what counts as an "addictive" platform or feature, and to the commissioner of mental health to prescribe the specific warning text.

"Users will not be able to bypass or click through the warnings," Hochul's office emphasized. (While her press release was Friday, the bill tracker shows the governor signed it Dec. 19 and Common Sense Media celebrated her signature of "our bill" Dec. 20.)

Ongoing Musk suit didn't stop James from making demands

Compelling speech from disfavored services, especially social media and pro-life counseling, has not gone well in court for New York, California and other politically blue jurisdictions in recent years, from the Supreme Court on down.

California abandoned its law forcing pro-life pregnancy centers to tell patients where they can get abortions and disclose that they don't perform or refer for abortions after SCOTUS found it likely unconstitutional in 2018. Delaware quickly gave up on its similar law this year following a First Amendment suit.

New York Attorney General Letitia James has flopped in her defense of both state law and her own enforcement practices around social media and pro-life counseling.

A judge blocked New York's law against "hateful conduct" on social media but refused to sanction James for subsequently pressuring platforms to censor constitutionally protected speech related to the Israel-Gaza war, deeming her letters "non-coercive" and "pursuant to at least one independent statutory authority" besides the hateful conduct law.

Her track record is worse on saber-rattling against pro-life pregnancy centers and activists, with courts rebuking James for trying to muzzle promoters of so-called abortion pill reversal and sidewalk counselors who seek to talk women out of entering abortion clinics. One judge even compared her efforts to the "Ministry of Truth" in Nineteen Eighty-Four.

The AG's office is facing a new lawsuit this month by school board members that alleges James threatened them from office for describing transgender students with sex-based pronouns and even for allowing debate on males in girls' facilities.

The onslaught of litigation hasn't dissuaded the former mortgage-fraud defendant from pushing the First Amendment envelope in collaboration with Gov. Hochul.

Despite an ongoing lawsuit by Elon Musk's X, James ordered social media platforms Oct. 2 to start disclosing in biannual reports "whether and how their existing policies deal with hate speech, racism, misinformation, and other types of content," as required by the Hochul-signed Stop Hiding Hate Act. X filed its opposition to James's motion to dismiss this month.

The Trump administration is newly divided on compelled speech for tech platforms, thanks to President Trump's nominee for antisemitism special envoy saying he favors mandatory labeling of supposed misinformation on social media and claiming the president and Secretary of State Marco Rubio are "firmly behind the efforts that I am doing."

The star plaintiff in the hateful-conduct challenge, First Amendment legal scholar Eugene Volokh, told Just the News that at "first glance" he doubts his eponymous group blog, The Volokh Conspiracy, is covered by the new warning-labels law but "I expect that other platforms that are covered will promptly challenge the law." (His blog's reader comments triggered the hateful-conduct law.)

Tech trade group NetChoice is a repeat challenger of laws regulating social media by both red and blue states but didn't answer queries Monday on how it might respond to New York.

SCOTUS gave NetChoice a second bite of the apple in litigation against Florida and Texas for their bans on content moderation but more recently allowed Mississippi's age-verification law to take effect over NetChoice's objections. NetChoice sued Virginia last month for its law requiring parental consent for children under 16 to use social media for more than an hour daily.

Regulators will fill in the many gaps later

Aside from Murthy's 2023 report for the Biden administration and leaked memos from Google, TikTok and Meta, New York's law is thin on sources for its claims justifying forced warning labels on social media, rattling off statistics about the "cumulative impact of these addictive design features" without attribution.

"Research shows that social media exposure overstimulates reward centers, creating pathways comparable to those of an individual experiencing substance use or gambling addictions — findings further bolstered by endless national surveys" of self-identified addicted teenagers who share their difficulties of limiting social media, the preamble says.

Common social media features draw "inspiration from the 'variable reinforcement schedules' produced by gambling slot machines that keep users pulling the lever … at periodic intervals for an outcome that could be intrinsically rewarding," it says.

The warning labels would resemble government-mandated messages that tobacco causes cancer and alcohol during pregnancy may cause birth defects, the preamble ends.

Much of the new chapter negotiated by Hochul outlines speed bumps users must endure to use social media, subjecting them to warning labels that cannot be bypassed at some as-to-be-determined frequency in each social media session.

Platforms could not put the labels "exclusively in the terms of service," obscure "the visibility or prominence" of the label such as by adding "extraneous text" to the government-required text, show the label "at a point" in the user experience or "for a duration" except as prescribed by the government. 

Also banned: "Deploying any other design feature or mechanism which intentionally serves to inhibit or subvert the purpose" of the law.

At least four commissioners would be involved in implementing the law, from writing the warning text based on their review of "medical and sociological research," to deciding how and when labels appear, then tweaking those government choices based on real-world user experience on an annual basis.

The law also tasks the AG with "promulgat[ing] such rules and regulations as are necessary to effectuate and enforce the provisions of this article," including granting exemptions and running a public complaint form for alleged violations by platforms. The AG can bring enforcement actions, seeking civil penalties of $5,000 per violation.