Internet CultureExplainer
New York's SAFE for Kids Act takes effect January 25: what changes for teen feeds and night alerts
From January 25, 2027, large platforms must confirm a New York user is an adult, or get parental consent, before serving algorithmic feeds or late-night alerts.
By Internet Culture Desk · · 5 min read
From January 25, 2027, social media platforms built around algorithmic feeds will have to treat users in New York differently depending on their age. Before a covered platform can show anyone in the state a feed that picks posts based on their data and past behavior, or send them notifications about that feed between midnight and 6 a.m. Eastern, it must either confirm through an age check that the user is an adult or, for a user it knows is under 18, obtain a parent's consent. That is the effect of New York's Stop Addictive Feeds Exploitation (SAFE) for Kids Act, which Governor Kathy Hochul signed in June 2024, and of the implementing rules that Attorney General Letitia James released in final form on July 28, 2026.
The Attorney General's office said the rules would be published in the State Register on July 29, 2026, and that the act takes effect 180 days later. Without consent, users under 18 keep access to the platform and its content, but their feeds show posts from accounts they follow or otherwise choose, in a set order such as chronological, according to the Attorney General's office.
Which platforms are covered
The law applies to what the rules call addictive online platforms. An addictive feed, in the rules' definition, is one that shows media shared or generated by users and recommends, selects or prioritizes it based on information tied to the user or their device, or on the user's earlier interactions with other users' posts. A platform is covered when 20% or more of the time its monthly active users spend on it goes to such feeds, measured over any six-month period of the previous calendar year that the platform chooses.
Some things do not count as an addictive feed: posts a user has expressly and unambiguously asked for, posts from creators a user subscribes to, and posts shared to pages or groups the user subscribes to. Smaller services are exempt if they have fewer than 5 million monthly active users or fewer than 20,000 monthly active users who are covered minors, meaning users in New York whom the platform knows to be under 18. Platforms whose primary user base is minors are not exempt.
| Rule | What it says |
|---|---|
| Start date | January 25, 2027 |
| Who is protected | Users in New York under 18 |
| Coverage threshold | 20% or more of monthly active users' time spent on addictive feeds |
| Exemption | Fewer than 5 million monthly active users, or fewer than 20,000 who are covered minors (known minors in New York), unless minors are the primary user base |
| Nighttime notifications | 12 a.m. to 6 a.m. Eastern |
| Age-check testing records | Kept for at least 10 years |
| Maximum civil penalty | $5,000 per violation |
How the age checks work
The rules allow three broad kinds of age assurance: estimation from a physical or behavioral feature, inference from other verified information, and verification with identification such as a government ID. The Attorney General's office gives uploading an image or video, or checking an email address or phone number against other information that reflects a user's age, as examples. It says platforms must offer at least one method that does not require a government ID.
Each method has to meet accuracy minimums. The rules cap how often a method may wrongly pass a minor as an adult: 0.1% of children aged 7 or younger, 1% of those aged 8 to 13, 2% at 14 and 15, 8% at 16 and 15% at 17. A method must also detect at least 98% of attempts to get around it. Platforms must have each method certified every year by an accredited third party, and keep the test results for at least 10 years. Data collected for an age check or for parental consent may not be used for anything else and must be deleted or de-identified immediately after use, except for basic records, such as the method used, the date of the check and the resulting age status, which platforms must keep for at least five years.
What parents and teens decide
A minor has to ask for the restricted features and agree to a parent being notified, and the parent must pass an age check before giving consent. Either can withdraw consent at any time. Platforms do not have to offer a consent route at all. A platform may not block a minor from using the service or searching its content because consent was refused, and beyond what compliance requires, it may not withhold or degrade features or raise prices for users or parents. It also does not have to show parents a teen's search history or interests to obtain consent, the Attorney General's office says.
Supporters and critics
James said the rules “set standards for age verification and parental control” so that tech companies do their part to keep children safe. Her office's release carried supportive statements from child-safety groups, including Common Sense Media and Fairplay, and from the Age Verification Providers Association and the Age Check Certification Scheme. The Ali Forney Center said LGBTQIA+ youth would still be able to reach affirming websites and crisis services under the rules.
NetChoice, which describes its work as making the internet safe for free enterprise and free expression, said the day after the rules came out that they amount to an unconstitutional censorship regime. It argues that platforms cannot know in advance whether they cross the 20% threshold, that annual testing and 10-year record keeping weigh most on smaller services, and that asking every user for identifying information creates a new target for data breaches. “New York doesn't need to replace parents,” said Amy Bos, its vice president of government affairs. NetChoice says it has won injunctions against similar laws in California and Arkansas, and it also notes that the age-assurance businesses quoted by the Attorney General would profit from the rules.
Enforcement and the road to the rules
The law lets the Attorney General sue to stop violations and to seek civil penalties of up to $5,000 per violation, among other remedies. Before the final version, the office published an advance notice of proposed rulemaking on August 1, 2024, with 60 days for comments, and made proposed rules public on September 15, 2025. Platforms that later lose their exemption get 180 days to comply the first time, and 30 days after that.
The rules apply only to users in New York, and platforms must take into account the reliable location information available to them, including technical data about a user's device, to work out who those users are. Families and companies with questions about their own situation can find the rules and contact details on the Attorney General's SAFE for Kids Act page. For the European Union's separate proposal, see our KIDS Act explainer.
Sources
- Attorney General James and Governor Hochul Release Final SAFE for Kids Act Rules to Protect Children Online Office of the New York State Attorney General, July 28, 2026
- SAFE for Kids Act Implementing Rules, Part 700 of Title 13 of the NYCRR Office of the New York State Attorney General, July 28, 2026
- New York's Final SAFE for Kids Rules Cement an Unconstitutional Censorship Regime NetChoice, July 29, 2026
Researched and drafted with AI assistance, then checked against the 3 sources above in a separate fact-check on November 14, 2026. Spotted an error? Tell us.