From Cigarettes to Scrolling: How the Big Tobacco Playbook Became Big Tech's Legal Nightmare

On April 14, 1994, seven men in dark suits raised their right hands before the House Subcommittee on Health and the Environment. Each was the chief executive of a major American tobacco company. Each swore, under oath, that nicotine was not addictive. The image became a symbol of corporate denial so potent that it outlasted the careers, companies, and even the lives of several of the men in the frame. Twenty-four years later, Mark Zuckerberg sat before the Senate Judiciary and Commerce Committees and declared that Facebook was a broadly positive force. In January 2024, he returned to tell the Senate Judiciary Committee that Meta already did enough to protect children. Between those two sets of testimony, separated by a generation, runs a legal strategy arc that lawyers on both sides of the current social media litigation know by heart. The question is no longer whether the parallel holds. The question is how far it goes.

The Forty-Year Wall

The American tobacco industry enjoyed a legal winning streak that stretched across four decades, and it did so not because the science was ambiguous but because the legal terrain favored the defense. From 1954 to 1973, plaintiffs filed approximately 150 lawsuits against tobacco companies. Not one succeeded. The industry did not need to prove cigarettes were safe. It needed only to argue that the plaintiffs had chosen to smoke, that the risks were publicly known, and that the causal link between smoking and any individual plaintiff's cancer was unprovable in that specific case. Against a lone plaintiff with lung cancer, these arguments were devastating.

The second wave of litigation, running roughly from 1983 to 1992, shifted to product liability theory. Plaintiffs no longer had to prove negligence. They had to prove the product was defective. But the industry adapted. Legal budgets that reportedly ran into the hundreds of millions of dollars annually by the early 1990s allowed tobacco companies to bury individual plaintiffs in discovery, motions, and appeals. The landmark case of this era, Cipollone v. Liggett Group, produced the first-ever plaintiff verdict against a tobacco company at trial in 1988, when a New Jersey jury awarded $400,000 in damages. But even that victory was partial: Rose Cipollone was found 80 percent responsible for her own injuries, and when the case reached the Supreme Court in 1992, the Court ruled that federal warning labels preempted some state law claims. The case was ultimately remanded and later dropped after the plaintiff's legal team, having spent more than $6 million over nearly a decade, could no longer afford to continue. The defendants had spent $40 million and never paid a cent.

The architecture of tort law, in other words, was built in a way that made individual litigation against a deep-pocketed defendant nearly impossible. A plaintiff needed years and millions of dollars to reach trial, and even a victory could be reversed on appeal. The defendant needed only to outlast the plaintiff. This is the structural lesson that matters for what comes next: the tobacco industry did not win because it was right. It won because it was rich and patient.

The social media litigation of the 2010s followed a strikingly similar early trajectory. Individual families sued Facebook, Instagram, Snapchat, and TikTok over their children's mental health. These cases were met with a defense that echoed the tobacco playbook almost verbatim: the platforms were legal products, used voluntarily, and the causal link between any particular child's anxiety or depression and any particular app was unprovable. Section 230 of the Communications Decency Act added a layer of legal armor that tobacco companies never had. For years, these individual suits went nowhere.

The Document Bomb

The event that shattered tobacco's legal fortress was not a scientific breakthrough or a sympathetic plaintiff. It was a document analyst named Merrell Williams. In 1994, Williams, working at a Louisville law firm that represented Brown & Williamson, leaked thousands of internal company documents. What they revealed was not merely that tobacco companies knew cigarettes were dangerous. They revealed that the companies had conducted private research confirming nicotine's addictive properties as early as the 1960s and had deliberately concealed that research from the public, from regulators, and from courts.

The scale of internal knowledge was staggering. A 1971 memo from Philip Morris researcher William Dunn stated bluntly: "The cigarette should be conceived not as a product but as a package. The product is nicotine." Brown & Williamson's own research division had documented nicotine's pharmacological effects in detail, while the company's public affairs division denied them. The gap between what the industry knew and what it said publicly transformed the entire legal landscape. Assumption of risk, the defense that had won case after case, required that the plaintiff knew the risk. If the manufacturer had actively concealed the risk, the defense collapsed.

The Williams leak triggered a chain reaction. Minnesota Attorney General Hubert Humphrey III filed suit against the industry in 1994 and won a discovery order that cracked open millions of additional internal documents. These eventually formed the basis of the Minnesota Depository, and later the Truth Tobacco Documents Library at the University of California, San Francisco, which grew to house more than 14 million internal industry documents. The documents did not merely help individual cases. They rewrote the public understanding of what the tobacco industry was: not a legal business selling a legal product, but an enterprise that had systematically lied about the nature of that product for decades.

The social media parallel arrived in September 2021, when Frances Haugen, a former Facebook product manager, copied tens of thousands of internal documents and delivered them to the Securities and Exchange Commission and to journalists at the Wall Street Journal. The documents, published as "The Facebook Files," showed that Meta's own researchers had found that Instagram worsened body image issues for a significant percentage of teenage girls, that the company's algorithms amplified divisive and harmful content because it drove engagement, and that internal proposals to address these problems were shelved because they conflicted with growth targets. The structural function of the Haugen leak in the social media litigation mirrors the Williams leak in tobacco litigation almost exactly: it transformed the legal question from "does this product cause harm?" to "did the company know it caused harm and choose to do nothing?"

The Third Wave

The legal breakthrough in tobacco came not from a better plaintiff but from a better plaintiff structure. In May 1994, Mississippi Attorney General Mike Moore filed the first state lawsuit against tobacco companies. The theory was elegantly simple: the state had spent billions in Medicaid funds treating smoking-related illnesses. The tobacco companies, which had known their product caused those illnesses, should reimburse those costs. The genius of the approach was that it sidestepped every defense that had protected the industry for decades. The state was not a smoker who had assumed the risk. The state was a payer who had been defrauded.

By 1997, all fifty state attorneys general had either filed suit or signaled their intention to join. The economics of the fight shifted overnight. Individual plaintiffs had been outspent and outlasted. Fifty state governments, backed by taxpayer-funded legal departments and contingency-fee private attorneys, could not be exhausted. The political dynamic shifted too: no attorney general wanted to be the last one standing with the tobacco industry.

In March 1996, the Liggett Group, the smallest of the major tobacco companies, broke ranks and settled with five states. The settlement was small, but its symbolic and legal significance was enormous: Liggett became the first tobacco company to settle smoking-related litigation, and the following year it signed a broader agreement with additional states in which it publicly acknowledged that smoking caused cancer and was addictive. The break signaled that the industry's united front was cracking.

This pattern is now repeating in social media litigation with remarkable fidelity. The individual family lawsuits of the 2010s gave way to institutional action. Thousands of cases were consolidated into Multi-District Litigation 3047 in the Northern District of California, before Judge Yvonne Gonzalez Rogers. State attorneys general began filing their own suits, reframing the question from individual harm to public cost. And the first break in the industry's ranks came when TikTok and Snap settled the K.G.M. bellwether case before trial for undisclosed terms, a move that echoed Liggett's 1996 decision so precisely that plaintiffs' attorneys openly cited the tobacco precedent in court filings.

The K.G.M. verdict itself, holding Meta and YouTube liable for product design that contributed to a minor's addiction, was the social media equivalent of Cipollone: the first time a jury said yes, this product caused this harm. In March 2026, New Mexico won a $375 million verdict against Meta in a state-level action, the first major AG victory. As of early 2026, more than 2,400 federal cases are pending in the MDL alone, with hundreds more in state courts and the number still climbing. The litigation mass is building in the same way, and at roughly the same pace, as tobacco's third wave.

The Settlement and What It Actually Changed

The Master Settlement Agreement was signed on November 23, 1998, between the four largest American tobacco companies and the attorneys general of 46 states. Four states had already settled separately: Mississippi, Florida, Texas, and Minnesota, for a combined total of approximately $36.8 billion. The MSA itself required tobacco companies to pay approximately $206 billion over 25 years to the remaining states.

But the financial terms were not the MSA's most lasting legacy. The agreement prohibited billboard advertising for cigarettes, banned sponsorship of sporting events, barred the use of cartoon characters in marketing (R.J. Reynolds had retired Joe Camel in 1997 under pressure), and required the creation and funding of the American Legacy Foundation, later renamed the Truth Initiative, which ran aggressive anti-smoking campaigns. The MSA also required the industry to fund the document depository, ensuring permanent public access to the internal records that had broken its legal defenses.

The results were measurable. The adult smoking rate in the United States, which stood at approximately 25 percent in 1997, fell to approximately 11 percent by 2023, according to CDC data. Youth smoking rates dropped even more sharply. The combination of advertising restrictions, public education campaigns, rising cigarette taxes (enabled by the political climate the litigation created), and social stigma accomplished what decades of surgeon general warnings alone had not.

But the tobacco industry did not die. It adapted. Philip Morris rebranded as Altria. R.J. Reynolds acquired Lorillard. The companies shifted investment to international markets where regulation was weaker, and they pivoted to new nicotine delivery systems. The rise of e-cigarettes and vaping, often marketed by companies with direct corporate lineage to the tobacco defendants, demonstrated that the MSA had changed the terms of the game without ending it. This is a distinction worth holding in mind as the social media litigation moves toward its own potential settlement.

The Map and the Territory

If the tobacco litigation arc is a map, then the social media litigation currently occupies a position somewhere around 1996 or 1997. The internal documents have been exposed. The first verdicts are in. The state attorneys general are filing. One defendant has broken ranks. The mass tort is consolidated and moving toward additional bellwether trials. The financial pressure is mounting.

The K.G.M. case was the proof of concept, the demonstration that a jury would accept the product-design theory of liability. The New Mexico verdict was the proof of scale, showing that state-level actions could produce nine-figure damages. The consolidation in MDL 3047 provides the mechanism for resolving thousands of claims without requiring each one to go to trial. The structural preconditions for a negotiated settlement are falling into place.

Plaintiffs' attorneys working the social media cases study the tobacco litigation not as history but as strategy. The parallels are not accidental. They are deliberate. The legal theories, the document discovery tactics, the sequence of individual suits followed by state actions followed by consolidation, all of it draws explicitly on the tobacco precedent. Several of the law firms leading the social media MDL cut their teeth on the tobacco cases a generation ago.

Where the Analogy Breaks

The tobacco parallel is the most powerful frame available for understanding where social media litigation is headed. It is also imperfect in ways that could prove decisive.

The first and most important difference is causation. Tobacco's harm was physical, measurable, and eventually undeniable. Lung cancer, heart disease, chronic obstructive pulmonary disease: these conditions could be observed, quantified, and linked to smoking through decades of epidemiological data. The dose-response relationship was clear. More smoking meant more risk. Social media's alleged harms are psychological: anxiety, depression, body dysmorphia, self-harm ideation. These conditions are real and serious, but their causal relationship to social media use remains scientifically contested. Serious researchers have found that the effect sizes are small, comparable in some studies to the effect of wearing glasses or eating potatoes. The K.G.M. jury accepted the causation argument, but the evidentiary standard in a mass tort settlement negotiation is different from the standard in a single bellwether trial. If the weight of scientific evidence remains ambiguous, the industry will use that ambiguity exactly as tobacco companies used the uncertainty of cancer science in the 1960s: as a delay tactic.

The second difference is economic structure. Cigarettes were a consumer product with a retail price. Taxing them, restricting their sale, and making them more expensive were all available regulatory levers. Social media is free to users. The product is funded by advertising revenue generated from user engagement, which means that the very design features alleged to be harmful are also the ones that generate the revenue. You cannot tax a free product. Restricting access raises questions about free speech that restricting cigarette sales never did. And the economic incentive structure is different: tobacco companies could survive higher prices because nicotine ensured demand. Social media companies depend on engagement metrics that are directly tied to the features under legal challenge.

The third difference is the constitutional dimension. Tobacco companies had no First Amendment defense. Their product was a physical substance, not expression. Social media companies argue that their algorithms constitute protected editorial judgment, that the decision to recommend one piece of content over another is a form of speech. Section 230 of the Communications Decency Act, enacted in 1996, provides platforms with broad immunity for third-party content. The K.G.M. case succeeded by arguing that the design of the product, not the content on it, was the source of harm. This product-design theory is a genuine legal innovation, but it has not been tested at the appellate level, and it has certainly not been tested at the Supreme Court. The constitutional questions surrounding algorithmic recommendation are unresolved in a way that the constitutional questions surrounding tobacco marketing never were.

The fourth difference is jurisdictional. The tobacco industry was overwhelmingly domestic. Its harms were concentrated in the United States, and the MSA was a US settlement. Social media companies operate globally. A settlement with American state attorneys general cannot bind operations in Europe, Asia, or Latin America. The European Union's Digital Services Act, China's time-limit regulations for minors, and Australia's outright ban on social media for children under sixteen represent parallel but distinct regulatory trajectories. A US settlement would be significant, but it would not be comprehensive in the way the MSA was for tobacco.

The Settlement That Has Not Yet Been Written

Despite these differences, the trajectory points toward a negotiated resolution. The financial exposure estimates for social media companies in the current litigation range from tens of billions to over $100 billion, depending on the scope of liability and the number of claims that survive summary judgment. Those numbers are large enough to focus corporate attention but not so large as to make settlement unthinkable. The tobacco industry paid more than $246 billion across all settlements and survived.

A social media settlement modeled on the MSA might include several components: mandatory age verification and design changes for minors, such as disabling infinite scroll, autoplay, and algorithmic recommendations for users under eighteen. It might require transparency in internal research, similar to the tobacco document depository. It might fund youth mental health programs, using the Truth Initiative model of industry-funded counter-messaging. And it might restrict certain marketing practices targeting minors, though the definition of "marketing" in a feed-based attention economy is considerably harder to draw than it was for billboard cigarettes.

The platforms may prefer to fight. Accepting design restrictions that limit algorithmic recommendation for minors would directly reduce the engagement metrics that drive advertising revenue. The economic incentive to resist is strong. But the economic incentive to settle is also strong: every bellwether verdict, every state AG victory, every leaked internal document increases the probability that the next jury will be larger, angrier, and less sympathetic. The tobacco industry resisted for forty years before calculating that settlement was cheaper than continued litigation. The social media industry is running the same calculation, and it is doing so with the tobacco precedent clearly in view.

The tobacco executives who raised their right hands in 1994 and swore that nicotine was not addictive were never prosecuted for perjury. The hearings, the lawsuits, the settlement, the advertising bans, the declining smoking rates: all of it took another four years to produce a resolution, and that resolution was a negotiated compromise, not a reckoning. The social media litigation is following the same arc, shaped by the same forces, constrained by the same legal structures, and driven by the same tension between public health and private profit. Where it ends will be determined not by the strength of the analogy but by the willingness of courts, legislatures, and the companies themselves to decide that this time, forty years is too long to wait.