Big Tech's Big Tobacco Moment? Not Completely — And Not for Everyone

Two groundbreaking trials and one landmark settlement against Meta over the past six months have led many to ask if this is finally the Big Tobacco moment for the social media industry. In March, a jury in New Mexico found Meta liable for misleading users about safety and enabling child sexual exploitation on their platforms, and a jury in Los Angeles found Meta and YouTube liable for designing addictive platforms that harmed a teenager’s mental health. In August, a New Mexico judge ruled that Meta created a public nuisance because its platforms’ design contributed to a worsening youth mental health crisis; and later that month, in a landmark settlement with 47 Attorneys General, Meta agreed to implement a number of design changes meant to make their platforms safer for children. Meta's total potential legal liability encompassing remedial abatement funds, civil penalties, and tort damages could reach up to $18 billion, and there are potentially hundreds of cases still to come.

Many are calling this Big Tech's "Big Tobacco moment" — the point at which internal documents, once buried, are read aloud in open court, and a company that spent years insisting it was doing its best is finally forced to answer for what researchers, journalists, and even its own employees warned about. In some respects, this comparison is apt. After years of feeling like accountability for this industry was elusive, these cases broke through and will have real financial and product design consequences. But this is only one piece of the larger puzzle of how to make a safer, more accountable online environment. 

There are myriad harms that stem from the ways our largest social media platforms are currently designed, deployed, and monetized. These cases shined a spotlight on many of the harms to children, but left untouched are issues like online scams and fraud; foreign interference in elections; online harassment; and the proliferation of hate and extremism. And herein lies the challenge: there is not a one-size-fits-all solution to the societal effects of social media. While these recent judicial victories certainly move the needle in the hard-fought road to accountability, realigning social media to be less divisive and harmful to broader society will require research-backed technical solutions that cannot always be addressed by the legal playbook we are witnessing at this moment.

Why these cases succeeded:

One of the major changes in these cases is that the claims made by plaintiffs were directly about product design features, not speech. Juries were shown features such as infinite scroll, autoplay, notification designs, and other behavioral tactics borrowed from the gambling industry, and heard from experts about how these design features harmed users. These features were designed to elicit a behavior on the part of the user that furthers the company’s own business goals, and it was that behavior change that contributed to the claims in these cases, as we argued in an amicus brief co-authored with the Electronic Privacy Information Center (EPIC). 

Juries in both New Mexico and Los Angeles were also shown internal documents, including Meta’s own research that acknowledged negative effects on children from certain design features, and internal communications with executives who still decided to pursue growth over safer design, even when shown the harms. At the heart of all these cases were questions of whether a company's product choices, made and re-made by executives who'd been warned of the consequences, caused foreseeable harm. The juries said yes. And following those first-of-their-kind results came the settlement with 47 other states, the District of Columbia, and several territories.

So is broader accountability on the horizon?

While we certainly applaud any efforts to design features and guardrails that actually will keep children safe online, the underlying architecture that incentivized dangerous design to begin with remains untouched in this moment: the engagement-based business model that requires keeping users on the platforms as long as possible, in service of selling personalized targeting tools to advertisers. 

It also does not change the fact that kids, as well as the rest of us, will still be served harmful content. Engagement-based algorithms will still feed teenage girls eating disorder videos, just as they will also still supercharge hate, antisemitism, and extremism. 

The core mechanism that amplifies hate on social media, high-intensity feed algorithms, does fuel a wide range of problems from scams and fraud to eating disorder content. Further, because these algorithms disproportionately amplify the most extreme content, they actually incentivize content creators trying to build large audiences to say ever more extreme things. However, these algorithms are not what Meta has agreed to change in the settlement. 

Quite intentionally, none of the agreed changes are ones that will impact what kinds of content Meta’s systems accelerate. Because once a claim hinges on the content itself, as opposed to design features or deceptive practices, it becomes an entirely different legal question that would have to pass First Amendment scrutiny. So while many have argued that the amplification of hate speech and extremism on social media has led to societal harm, this problem will be much more complicated to solve.
Could a future court case, or US law, lead to the kinds of design changes that would break this cycle of acceleration of the most hateful content? We suspect the answer is no. Here is the simple fact that complicates the quest for a less toxic, hate-fueled online environment: unlike the unsafe product designs and deceptive trade practices that were at issue in these recent cases, hate speech is not illegal in the United States. Even if a company amplifies, incentivizes, or profits from hateful rhetoric, this is still fiercely protected. If Mark Zuckerberg, or any other tech CEO, wants to run a platform that amplifies hate, then the First Amendment strongly protects their right to do so (as long as it doesn’t also advocate for imminent violence). 

What we need is an equivalent reckoning for the corrosive effects on our social fabric of algorithms that amplify hate and social networks that favor extremism and incentivize users further and further away from a common ground. That challenge will require a different playbook, one that recognizes that this path through the court system around the issue of harms to kids from design patterns will not be the path forward for combating hate. However, there are some things we can learn from the long process these cases have gone through.

A key ingredient to accountability: independent research

A key factor in these cases’ success was the advance of research to better understand the harms kids are experiencing, and the link between those harms and the product designs in question. Whistleblowers who took risks to expose both the internal research and executive decision-making, expert testimony, and academic research were absolutely key to establishing plaintiffs’ claims and in justifying why design changes should be part of any remedy agreement. (In full disclosure, some experts who testified in these cases are members of our lab, and our research was introduced as evidence.)

Lawsuits and new laws may not be the way forward for combating hate speech, but independent research will continue to be key. The fight against extreme hate is going to be waged in the public square, not the courts. Facts and knowledge are just as valuable in that fight. When the public finally understands how social media itself is contributing to an environment they want changed–as is happening with kids’ safety–only then will the public truly push for change. In the case of hate speech and other toxic, speech-related issues emanating online, public demands for change can move the needle if, and only if, that demand results in economic harm to the company through less engagement, or more importantly, less advertising.

Better diagnoses of the technical problems, and research-driven development of technical solutions won’t, by themselves, solve the problems of hate online. But to advocate for solutions to problems caused by technology, we need to understand how the technology itself is working and know what the technical solutions would be that we are advocating for. 

There is no silver bullet, but change is still possible

So: is this Big Tobacco's moment for social media? For child safety, we are finally on that road — verdicts, discovery, and now enforceable settlements are exactly the machinery that eventually reshaped the tobacco industry. But tobacco didn't just get "safer" for kids while continuing to poison everyone else. If we stop at child safety, we will have won an important battle while leaving the deeper architecture of algorithmic feeds designed to maximize engagement — the machinery that also radicalizes, polarizes, and puts targets on people's backs — almost entirely untouched. 

However, these child safety cases solidified for the public that these companies could, if they chose to, design safer platforms. They also reinforced that we cannot just take the companies at their word and allow them to grade their own homework. That’s why commitments to transparency and researcher access are also key to keeping the public informed and holding companies accountable for the promises they make. And while the current legal playbook will not be the path to fixing online hate, harassment, extremism, and other terrible content-driven issues, we at least know that we can demand more of these companies, especially when we are armed with the knowledge of how they work and how they are impacting our communities.