Social Media Warning Labels: Will They Actually Work?
The Surgeon General wants them. California is mandating them. A Stanford study tested them on 1,000 young people. Here’s what the research says warning labels can do, what they can’t, and what you need alongside them.
Social media warning labels are no longer hypothetical. California will require them on every major platform starting January 1, 2027. Minnesota already does. Forty-two state attorneys general want Congress to make them federal. The question isn’t whether they’re coming. It’s whether they’ll actually change anything.
The short answer: warning labels shift how people think about social media. A 2026 Stanford study confirms that. But there’s no evidence yet that they change how people use it. The tobacco comparison everyone loves to make is more complicated than it sounds. And if you’re waiting for a government label to fix your scrolling habit, you’ll be waiting a long time.
What the Surgeon General Actually Said
In June 2024, U.S. Surgeon General Vivek Murthy published a New York Times op-ed calling for tobacco-style warning labels on social media platforms. His argument was blunt: “Social media has not been proven safe” for young users.
The call wasn’t impulsive. His May 2023 advisory had already described social media as a “profound risk” to youth mental health, citing research linking heavy use to doubled anxiety and depression risk. Teens spending more than three hours daily on social media faced significantly higher rates of mental health problems. By 2024, he’d moved from concern to policy: put a warning on the box, the way we did with cigarettes.
There’s a catch. The Surgeon General can’t do this alone. It requires an act of Congress. And Congress, as you might have noticed, doesn’t move fast.
The States Aren’t Waiting for Washington
While Congress debates, states have started writing their own rules.
California’s AB 56, signed by Governor Newsom in October 2025, is the most aggressive. Starting January 2027, social media platforms must display health warnings to users under 18. The warnings cite the Surgeon General’s advisory and state that social media is “associated with significant mental health harms” for young users.
The design is deliberate. Warnings appear on first daily access, again after three hours of active use, and then hourly. They last at least 10 seconds initially and 30 seconds on subsequent displays. Platforms can skip users they can reasonably determine are over 17.
Minnesota was first. Texas, Ohio, and Pennsylvania have proposed similar bills. The momentum is real, even if the legal challenges are already forming. TechNet, the industry trade group representing Meta, Google, and others, opposed AB 56 on First Amendment grounds, arguing it forces platforms to carry “government-mandated labels on user-generated speech.”
Sound familiar? The tobacco industry made the same argument. They lost.
The Tobacco Precedent: What Warning Labels Did to Smoking
Everyone compares social media labels to cigarette warnings. Fair enough. But the tobacco story is more instructive than most people realize.
Canada introduced graphic warning labels on cigarette packages in 2001. A nine-year study published in Tobacco Control compared Canadian and U.S. smoking rates before and after the labels went up. The result: a 12 to 20 percent decrease in the number of Canadian smokers over that period. Researchers estimated that applying the same approach to the U.S. would reduce smokers by 5.3 to 8.6 million.
But here’s what the comparison leaves out. Tobacco labels didn’t work alone. They were part of a package that included advertising restrictions, tax increases, public education campaigns, and smoking bans in public spaces. No one credible argues that a label on a pack of Marlboros single-handedly made people quit.
Labels did something specific: they changed the cultural framing. Smoking went from glamorous to disgusting. The graphic images of diseased lungs weren’t just information. They were stigma. And stigma works.
Key insight: Tobacco warning labels reduced smoking by 12–20% in Canada over nine years. But they were part of a broader campaign, not a standalone fix. Social media labels face the same limitation.
What the Stanford Study Found
In September 2026, Stanford Medicine published the first serious test of social media warning labels in JAMA Health Forum. Researchers led by Anna Grummon ran an online experiment with about 1,000 teens and young adults.
They tested three types of messages: an explicit health warning linking social media to mental health harms, a time-break reminder (similar to the one TikTok already uses), and an unrelated seatbelt-safety message as a control.
The health warning won convincingly. Messages about mental health harms, depression, and anxiety were rated as most effective by participants. Nearly every health-focused warning outperformed the screen-time break reminder.
Participants said the warnings would make them use social media less. They also scored higher on knowledge of social media risks after seeing the labels.
Now for the caveat that changes everything: the study did not measure actual behavior. Nobody’s phone was monitored. Nobody’s screen time was tracked. Participants said they’d change. Whether they actually would is a different question entirely.
If that gap between intention and action sounds familiar, it should. It’s the same reason people set screen time limits and blow past them by noon.
What Warning Labels Can Do
Labels aren’t useless. They serve three functions that research supports.
They change the cultural narrative
Before tobacco labels, smoking was normal. After, it was a health risk you chose to accept. Social media is still in the “normal” phase. A government-mandated warning reframes it as something with known risks. That cultural shift matters, especially for parents trying to set boundaries without feeling like the bad guy.
They increase risk awareness
The Stanford study confirmed this directly. People who saw health warnings knew more about social media risks afterward. Many young users genuinely don’t know that heavy social media use correlates with depression, anxiety, and sleep disruption. A label at least puts the information where they can’t miss it.
They signal governmental seriousness
When the Surgeon General puts a warning on something, it tells the public that the evidence has passed a threshold. It tells researchers that funding will follow. It tells platforms that regulation is coming, which can incentivize voluntary design changes. Meta and TikTok have already added (weak) screen time features partly in anticipation of this pressure.
What Warning Labels Can’t Do
Here’s where it gets honest.
A label that says “social media is associated with mental health harms” does not create friction. It does not make your phone less rewarding to pick up. It does not interrupt the dopamine loop that keeps you scrolling. It tells you the stove is hot. It doesn’t move your hand.
The Stanford study found that TikTok’s existing screen-time break reminders performed worse than the health warnings. That’s because those reminders are the platform equivalent of a bartender saying “you sure you want another?” while refilling your glass. The incentive structure hasn’t changed. The platform still profits from every second of your attention. The attention economy runs on engagement, and a 10-second warning label doesn’t alter the math.
Northeastern University researchers flagged this problem directly: knowing something is bad for you doesn’t automatically change compulsive behavior. Every smoker who ever read “SMOKING KILLS” and lit up anyway proves the point.
What Actually Reduces Social Media Use
If labels are step one, what’s step two? The research points to friction. Not information about harm. Not awareness of risk. Friction.
Studies consistently show that making phone use slightly harder or less rewarding cuts actual screen time in ways that labels and timers don’t. Here’s what works:
Grayscale mode
Four peer-reviewed studies show grayscale cuts phone use by 20–38 minutes daily. Color is one of the primary hooks that makes social media visually rewarding. Remove it and the compulsive pull weakens. Go Gray makes this a one-tap switch from Control Center, the Action button, or Back Tap.
App-specific nudges
Blanket time limits treat every app equally. But your problematic patterns probably live in two or three apps. Go Gray lets you pick the apps that pull you in and nudges you after 30 minutes and again after an hour on them. That’s targeted friction where it matters.
Removing persuasive design elements
A study cited in the persuasive design literature found that eliminating engagement-maximizing features would cut phone use by 37%. Infinite scroll, autoplay, and like counts are designed to keep you hooked. Until platforms are forced to remove them, you need to counteract them yourself.
Environment design
Phone-free zones in your bedroom, at meals, and during work create structural barriers. You’re not relying on a label to remind you. You’re removing the phone from the context where it does the most damage. A nightly phone curfew adds 40 minutes of sleep for those who stick with it.
Labels Plus Friction: The Combination That Works
The tobacco analogy is actually useful here, just not the way most people use it.
Warning labels didn’t kill smoking. Warning labels plus smoking bans plus tax increases plus cessation programs did. The label was the awareness layer. Everything else was the friction layer.
Social media needs both. California’s law puts the awareness layer in place. But the friction layer is up to you. No government label will make Instagram less addictive. No warning will interrupt your doomscrolling at midnight. That takes a personal tool.
Go Gray is built for exactly this gap. Turn on grayscale with one tap from Control Center when you want to focus. Set nudges for the apps that pull you in. Use 30-minute color breaks when you actually need color for a photo or a map, and it switches back to gray by itself. You get the friction that labels can’t provide.
Warning labels are a good start. A society that puts health warnings on social media is telling the truth about what these platforms do to people. But the label doesn’t fix you. You still have to put the phone down. And that’s the part where friction beats information, every single time.
Frequently Asked Questions
Will social media have warning labels?
Do warning labels on social media actually work?
What does the Surgeon General say about social media?
Which states require social media warning labels?
What can I do besides wait for warning labels?
References
- Murthy, V. (2024). “Why I Am Calling for a Warning Label on Social Media Platforms.” The New York Times, June 17, 2024. U.S. Surgeon General’s call for tobacco-style warning labels.
- Grummon, A. et al. (2026). Perceived Effectiveness of Health Warning Labels for Social Media. JAMA Health Forum, September 2026. Stanford Medicine randomized experiment with approximately 1,000 teens and young adults.
- California Assembly Bill 56 (2025). Signed by Governor Newsom, October 2025. Requires social media platforms to display health warnings to users under 18, effective January 1, 2027.
- Hammond, D. et al. (2013). The impact of cigarette warning labels and smoke-free bylaws on smoking cessation. Tobacco Control. Nine-year comparison of Canadian and U.S. smoking rates showing 12–20% reduction in Canadian smokers.
- Murthy, V. (2023). “Social Media and Youth Mental Health.” U.S. Surgeon General’s Advisory, May 2023. Described social media as a “profound risk” to youth mental health.
- National Association of Attorneys General (2024). Bipartisan letter from 42 state attorneys general urging Congress to enact social media warning labels.
Labels tell you the problem. Go Gray fixes it.
One-tap grayscale, app nudges, and 30-minute color breaks. The friction that warning labels can’t provide.
