
Last week, a Los Angeles jury found Meta and YouTube negligent for designing addictive features that harmed a young woman’s mental health. Meta was ordered to pay $4.2 million. YouTube, $1.8 million. Snap and TikTok settled before the trial began. The headline in every publication was some variation of the same idea: this is Big Tech’s Big Tobacco moment.
We say maybe. But the comparison, as satisfying as it feels, is doing more work than it should. Tobacco and social media share some structural DNA — a product that harms people, a corporation that knew, an army of lawyers — but the differences matter enormously. And they cut in both directions. Some make this fight harder for plaintiffs than tobacco ever was. Others make it far more dangerous for the companies.
If you want to understand where this actually goes, you need to hold both truths at once.
Where the Tobacco Analogy Works
The legal architecture here is clever. For years, social media companies hid behind Section 230 of the Communications Decency Act, which shields platforms from liability for content their users post. The plaintiffs in this trial didn’t attack the content, they attacked the design. Infinite scroll, autoplay, notification cascades, algorithmic recommendations calibrated to keep a nine-year-old watching until her dopamine receptors look like a lab rat’s.
By reframing the claim as product liability — a defective design, not defective speech — the plaintiffs navigated around Section 230 entirely. The jury was asked a simple question: did these companies design a product they knew was addictive, and did they sell it to children anyway? The answer was yes, twice.
That’s broadly the same move that broke tobacco. For decades, cigarette companies won lawsuits by arguing that smokers assumed the risk. Plaintiffs only started winning once the legal theory shifted to fraud and concealment — the companies knew nicotine was addictive and cancer-causing and hid the data. The “what did you know and when did you know it” framework is what ultimately produced the 1998 Master Settlement Agreement and its $206 billion in damages.
Social media plaintiffs are now finally running the same play: internal documents showing that Instagram’s own researchers found the platform worsened body image in teenage girls, and that the company buried the findings. The jury cared about that. Juries always care about that.

Where the Analogy Falls Apart
Here’s where things get uncomfortable for the optimistic “just like tobacco” crowd. Cigarettes cause cancer. The causal chain is physical, measurable, and lethal: nicotine enters the bloodstream, carcinogens accumulate, tumors form, people die. Decades of epidemiological data make the link incontrovertible. No serious scientist disputes it.
Social media’s harm mechanism is behavioral. The claim is that design features create a pattern of compulsive use that worsens anxiety, depression, and body dysmorphia in adolescents. There certainly is research supporting this: adolescents who use social media more than three hours per day face double the risk of depression symptoms. But it’s still epidemiologically messier. There are confounding variables everywhere. The DSM-5 (standard classification of mental disorders used by mental health professionals, published by the American Psychiatric Association) doesn’t recognize social media addiction as a clinical diagnosis. Only one behavioral addiction, gambling disorder, currently holds that status.
There exists an ambiguity which is what appeals courts will seize on. Trial juries are human beings — many of them parents — who feel the evidence viscerally. They have seen the effects of social media on their children. Appellate panels are trained to feel nothing. The question on appeal will be whether the scientific foundation beneath the product liability theory is strong enough to survive Daubert scrutiny, and honest legal scholars are split on the answer.
And then there’s the First Amendment problem that tobacco never faced. Cigarettes are a physical product. Social media platforms are, in part, speech services. Meta and YouTube will argue that their recommendation algorithms curate lawful expression, and that regulating how that expression is delivered amounts to regulating speech itself. Whether that argument succeeds is an open question — but it’s a question that Philip Morris never had to litigate.

The Insurance Bomb Nobody’s Talking About
If you’re looking for the single most important development in this story, skip the verdict. Read the insurance ruling.
In February, a Delaware judge ruled that Meta’s insurers have no duty to defend the company in social media addiction lawsuits. The allegations describe deliberate, intentional conduct — not the kind of accident that commercial general liability policies are designed to cover. If Meta knowingly designed an addictive product and knowingly marketed it to children, the insurers argue, then Meta was knowingly breaking the law, and no insurer is obligated to backstop that.
This matters more than the $6 million verdict. Meta can absorb $6 million without its CFO even knowing about it. Meta cannot absorb thousands of uninsured lawsuits. The company currently faces roughly 3,400 individual suits, 1,700 school district complaints, and actions from 43 state attorneys general. If the insurance ruling holds, and if the design defect theory survives appeal, Meta is looking at a litigation tab measured in tens of billions, without an insurance backstop, while simultaneously trying to spend $65 billion a year on AI infrastructure. Philip Morris had insurance. Meta might not. That’s the difference that should keep Zuckerberg’s lawyers up at night.

The Real Tobacco Parallel: Who Actually Pays
The most under-discussed element of the tobacco comparison is distributional. The 1998 MSA didn’t destroy tobacco companies. Philip Morris still exists and has cashflows. People still smoke. What the settlement did was redirect revenue — roughly $9 billion per year flows from tobacco companies to state governments. Some of that money funds smoking prevention, but most of it doesn’t. The states, in many cases, securitized the future payments and spent the money on roads and pension shortfalls.
Social media regulation will likely follow the same pattern. The companies won’t disappear. Instagram won’t shut down. What will change is the cost structure. Design modifications — time limits, disabled autoplay for minors, algorithmic throttling — will reduce engagement metrics, which will reduce ad revenue, which will reduce margins. The question is who absorbs that margin compression. If history is any guide, shareholders take the first hit, advertisers adjust next — and whether the children these lawsuits are meant to protect see meaningful benefits at all is a much less certain question.
That’s the very cynical read. Here’s the less cynical one: the tobacco litigation, for all its flaws, changed behavior. Smoking rates in the U.S. dropped from 42% of adults in 1965 to roughly 11% today. The litigation didn’t do that alone — public health campaigns, warning labels, indoor smoking bans all contributed — but the legal reckoning created the political and financial conditions that made those interventions possible. A generation of Americans grew up understanding that cigarettes were a product designed to addict them, sold by companies that lied about the risk.
Social media may be headed for the same generational shift. The research is accumulating. The juries are listening. Forty percent of Americans aged 18 to 22 say they feel addicted to social media, and unlike previous generations of smokers, they’re saying it while they’re still using the product. That kind of cognitive dissonance, weaponized by trial lawyers and validated by jury verdicts, is how mass tort litigation gathers escape velocity.

Big Tech(bacco)?
The design defect theory is strong, the internal documents are damning, and the jury has spoken. The pending lawsuits number in the thousands, and the insurance shield just cracked. All of that tracks the tobacco playbook.
But the First Amendment complication has no tobacco precedent. The scientific causation is softer. The appellate road is long. And the damages so far — $6 million against companies worth a combined $3 trillion — are a rounding error on a rounding error.
Ultimately, this is bigger than people think and slower than people hope. The tobacco reckoning took 30 years from the Surgeon General’s first warning to the MSA. The opioid reckoning took 20. Social media went mobile in 2013. We’re 13 years in. The clock is running, but we’re probably still in the middle innings — not at the finish line.
What I’m watching: the appeal. If the Ninth Circuit validates the design defect theory, every plaintiff’s attorney in America will file by morning. If it doesn’t, this becomes a state-by-state regulatory fight that drags on for a decade.
See ya folks, and stay curious.
—J&E
Disclaimer: This post is for informational and editorial purposes only and does not constitute legal or investment advice. All data sourced from court filings, CDC, Pew Research, Gallup, DemandSage, and public reporting. Past performance is, as always, a liar with good posture.