New York bans 'addictive feeds' for teens | Hacker News Reader

New York bans 'addictive feeds' for teens

theverge.com

94 pointsruddct108 comments

The “definitions” section of the bill is a lot more decent than I expected it to be.

Page 3 of the bill pdf:

https://legislation.nysenate.gov/pdf/bills/2023/S7694A

> [The definition for Addictive Feed applies] unless any of the following 18 conditions are met, alone or in combination with one another:

> ... (b) the recommendation, prioritization, or selection is based on user-selected privacy or accessibility settings, or technical information concerning the user's device;

And also:

> It shall be unlawful for a covered operator to provide an addictive feed to a covered user unless:

> (b) the covered operator has obtained verifiable parental consent to provide an addictive feed to a covered minor.

"Hey, kid, activate the Dynamic Feed setting and get a cool badge on your profile!" Or: "Type your parent or guardian's email here; we'll tell them all about how we work to provide you with age-appropriate content, and if they click to upgrade your Dynamic Feed, you'll get a special hat for your avatar!"

This seems rife for abuse in practice, but I'd rather this be the case than have overly-broad definitions.

(Not a lawyer, this is not legal advice!)

That sounds like: lets appear that we are doing something, but make it easy enough to go around so it doesn't hit our bottom line.

I think it realistically needs an escape hatch. For example some sports / activity groups organise exclusively through Facebook. If Facebook solves the issue by not allowing NY teens any access, then parents may want to override that for that specific purpose. Or the teen may be working on something that integrates with a service providing the addictive feed - it would make sense to allow.

And honestly, knowledge is half the battle. People seeing "Addictive Feeds" as a dangerous product that needs regulating is worth more than the actual regulations.

I think they're trying to make it go the way of smoking:

- general consensus of harm

- warning labels

- start applying pressure to phase out

Additionally, why do they allow to call it "Dynamic Feed" and not for what it is: "Addictive Feed"? Seems like doublespeak.

My implication is that Facebook will solve this, not by disallowing access to NY teens, but by incentivizing NY teens and their parents to opt into the exact type of addictive feed that they currently have, and in the meantime falling back to non-user-specific recommendations. Also, it appears that a platform can still show content from users/channels that a minor user has explicitly subscribed to, without that counting as an addictive feed.

It's a very de-fanged law, but it's an iterative step in the right direction.

Now, with improved >Addictive Feeds< you'll get cool blue messaging bubbles instead of plain green ones!

Please ban it for me too thank you

HN is my most addictive social media, should this site be banned? :)

I see the name "dang" discussed here quite a bit; it seems to be an internal code name for a highly effective algorithm for surfacing relevant content. Whatever it is, it seems to be far more intelligent than any other automated system I've seen. Has HN somehow created a sentient recommendation system?

No, it's just a minced oath, as in "dang, that was a good moderation call".

The algorithm is actually quite old. While dang is the result of numerous modern innovations, the approach has roots in a 1770 technique due to Johann Wolfgang Ritter von Kempelen de Pázmánd, who is perhaps better known for his work on speech synthesis.

While you may see other users talking about "him" in a personal manner, any appearance of sentience is, I assure you, entirely ineluctable.

> ineluctable

Love to see this word used! Thank you once again HN for being a great place where I can smile every day.

Karma probably should be

You can lie on your birth date in your profile.

This is obviously unconstitutional and it'll be shot down in the courts before it takes effect since this is their version of KOSA among other First Amendment issues.

Some context here: https://www.techdirt.com/2024/06/11/nys-safe-for-kids-act-a-...

What is unconstitutional about it?

It's content-based discrimination, which means it goes straight to strict scrutiny.

How is it discriminating content?

It’s not putting restrictions on content that can be shown.

It's discriminating between social media companies purely on the basis of their speech--the recommendation algorithms they choose to use.

If the algorithm constitutes speech by the social media companies, doesn’t that remove their common carrier status?

If the speech is theirs, they are responsible for its content.

> If the algorithm constitutes speech by the social media companies, doesn’t that remove their common carrier status?

They aren't common carriers and never have been.

Good god, not that zombie lie again!

[1] Age veriifcation

[2] Content discrimination

[3] Has the unfortunate side effect of restricting LGBTQIA+ youth and adults' speech

There's a reason why most of these bills are legally challenged.

I don’t understand #3.

First Amendment doesn't cover things that cause direct harm or speech to minors. In this case, you can draw a straight line between social media usage and an increase in teen suicides, and keeping them on it for long periods is detrimental to their overall mental health especially when they don't have fully developed impulse control or are not legally expected to. You're not allowed to advertise gambling products, alcohol and tobacco to minors under the First Amendment. You can't intentionally show porn to a minor without going to jail in many cases. (18 U.S. Code § 1470 - Transfer of obscene material to minors)

You want to get adults addicted and express yourself to them, so be it, but the legal distinction on limitations on what you can express to minors is well documented.

Just a reminder that getting adults addicted to stuff is not always legal either.

> You can't intentionally show porn to a minor

Anyone can go to pornhub and see a landing page full of highly explicit sexual images without any kind of age verification whatsoever. How is this legal, if what you say is true? Is it just that they don't know it's a minor on the other side of the screen, so they can say it's not "intentional"?

Intent is when someone mentally commits to an action

Pornhub home page has a very clear warning before you get to anything explicit:

    This is an adult website

    This website contains age-restricted materials including nudity and explicit depictions of sexual activity. By entering, you affirm that you are at least 18 years of age or the age of majority in the jurisdiction you are accessing the website from and you consent to viewing sexually explicit content.

    I am 18 or older - Enter 
    I am under 18 - Exit

I don't see that. And no, it's not because I have cookies from them.

Maybe it's geo-location dependent?

Try it in a private browser, without any cookies? May also be an ad blocker

Try logging out.

> Anyone can go to pornhub and see a landing page full of highly explicit sexual images

Not in an increasing number of states where age verification is now required.

Seems like they're getting around the complaints about the previously overturned laws by just going after the feed order while not banning access to the underlying content via conventional means.

> This is obviously unconstitutional

"or abridging the freedom of speech, or of the press"

Well, is social media "speech" or "press"?

Television and radio stations need to abide by FCC guidelines, and in general, certain topics and words aren't allowed in the daytime.

Another example: Tobacco advertising is severely curtailed, *especially advertising targeted towards minors."

In this case, what isn't restricted is social media platforms ability to express ideas or otherwise function as "press." Instead, what's restricted is the addictive nature and activities that cause psychological harm.

> Well, is social media "speech" or "press"?

whichever benefits them at that point in time

> Well, is social media "speech" or "press"?

Speech.

> Television and radio stations need to abide by FCC guidelines

As far as the content restrictions, they only need to do so if they are broadcast channels (or cable rebroadcasts of broadcast channels), and FCC is constitutionally prohibited from regulating speech on cable-only channels. The argument for why they can do broadcast is the "compelling government interest" in policing the inherently limited broadcast spectrum, which (quite frankly) is pretty shaky precedent if you ask me.

> Another example: Tobacco advertising is severely curtailed

Advertising is commercial speech, which the government has much more powerful abilities to restrict than expressive speech.

There's a pretty strong argument that an algorithmic feed optimised for engagement is commercial speech.

How is that feed "proposing a commercial transaction"?

It’s showing you content in exchange for your attention to their ads?

So is every newspaper.

I don't think you understand what "speech" and "press" are if you argue that the algorithm itself, or the act of running a social network, is protected speech.

In the case of the FCC, there's a clear difference between a radio signal, and the message it carries.

If you think the FCC is unconstitutional, I don't think you are able to make an argument with merit regarding social media.

You have the right to say want you want, but that doesn’t mean parents or the state have to let you show it to teenagers.

The bill sponsors are senator Andrew Gounardes and assemblywoman Nily Rozic. Rozic told CBS news: "Jewish parents are being told to remove social media apps completely from their phones because there is no way to actually prevent what their children will see."

> This is obviously unconstitutional

I also wonder if this is a first salvo to get users to submit IDs before using online services. How do you determine if a user is a minor or an adult without some form of ID, right?

Exactly, since age verification is going to be required which means collecting a LOT of personally idenificable information (PII) by ways of ID which govenrnments don't have a good history in securely storing that information.

Some more context:

Hochul told a reporter, “we’ve checked to make sure, we believe it’s constitutional.” And, that’s just laughable. Checked with whom? Every attempt I saw to call out these concerns was brushed off as “just spewing big tech’s talking points.”

The Constitution is not a “big tech talking point.” What the actual research shows is not a “big tech talking point.”

https://www.techdirt.com/2024/06/21/today-we-save-our-childr...

Another case of the political use of medical terms in completely unsupported ways for political goals. There is no medical or scientific support for the use of the word "addiction" in this context. It's a meme that's become more dangerous than the problem it imagines. It's not in the DSM 5 or ICD10 and not because they haven't addressed it in committee. They have, as recently as 2020 and found no support for it. And if you say, "gambling" well, yes, that's "gambling disorder" not "gambling addiction" and yes, words matter. Especially in legislation.

>“We’ve checked to make sure, we believe it’s constitutional.”

It may be constitutional but it certainly isn't scientific. It's closer to the for-profit use of medical ideas in "anti-gay" "de-patterning" camps and the like. Except backed by people with firearms and a tendency to use them without consequence.

> Another case of the political use of medical terms in completely unsupported ways for political goals. There is no medical or scientific support for the use of the word "addiction" in this context.

Say the language is fixed up replacing addiction with disorder, ... Any other objections?

That would be a great first step.

My other big objection is that this bill as written applies to all people hosting websites instead of being restricted to just incorporated persons like corporations and institutions like the EU's Digital Markets Act.

If it were modified to only apply to incorporated entities then it'd be a solid bill doing net good without any significant violation of individual rights. But as is it applies to all human people that run websites and that makes it a net negative because of the violation of the rights of normal human people.

Eh, this feels like a bad-faith argument. I think most of us on this site first-hand feel that obsession to update a feed that seems to control us more than we control it.

I don't need the DSM's permission to notice that.

Whether the ban will work out, open question. But pretending there's no addiction here doesn't pass the smell test.

I have never felt any compulsion to update a feed. I would like some actual numbers of people that have a debilitating relationship with social media before we pass a bunch of laws restricting it.

Lots of people have never experienced depression but don't feel the need to claim that means it's not real.

I didn't make any sort of claim like that. The parent said most people have social media compulsion, and I doubt those claims. If you are going to ban something from an entire population of people, why should they not study the actual prevalence of harm.

Even if it may feel universal that is likely not reflecting the actual distribution of the perception. (ref: Inferring the Popularity of an Opinion From Its Familiarity: A Repetitive Voice Can Sound Like a Chorus https://www.apa.org/pubs/journals/releases/psp-925821.pdf)

Additionally, feelings are fine for personal behavior but legislation requires a higher level of evidence. I really do believe what I am saying: addiction is an inappropriate concept to apply here since incentive salience is not being directly hijacked. The types of legislative responses to social problems of addiction (like to cocaine) are not appropriate or justified in this context.

To be clearer: enjoyable things with intrinsic value are being targeted in this context and those things are enjoyed. While addiction involves uncontrollable reptition of things without intrinsic value which become wanted due to the system for wanting being activated directly. Stimuli on screens do not do this. Drugs do. That's why it's gambling disorder and drug addiction. That extra layer of abstraction through the senses makes all the different.

> Additionally, feelings are fine for personal behavior but legislation requires a higher level of evidence.

This I can agree with -- but I think it would trivial to get that evidence by asking 10,000 teens if they agree with the following statement "Some of my social media apps feel addictive -- they don't bring my enjoyment but I can't seem to stop using them."

I'd bet about half of teens agree with that.

You can get addicted to anything that gives you little dopamine hits and stimulating reward centers the same way gambling, etc does. Dark patterns are well documented. Medical science and multiple studies involving brain scans during social media use and lighting up those areas of the brain back this up. The DSM hasn't been redefined but there's recognition of it as a problem and Internet/Gaming (digital media) addictions being added to the DSM is being discussed and there's no argument these platforms are designed by psychologists to use our own brain chemistry against us and create these addictive behaviors. How this is done is well documented and its exactly what the law lays out in it's defining of what constitutes an addictive feed.

Source: Just search HN, there's been so many posts related to this.

There is no requirement for laws to use words with the same meaning as they’re used in one specific niche publication. “Addictive” is a normal English word; it doesn’t exclusively belong to psychiatry.

Addiction was a meaningful word long before the DSM

Correct, which is why the DSM doesn’t get to define what it means for all English-speakers.

They're not? They're Defining what it means in a medical setting so we can all have a standard set of evaluating principles to work and diagnose from in order to ensure a standard level of care is achieved across all those under their care.

I don't agree with DSM decisions but I don't have to in order to see this basic nuance about their purpose.

Sure, but this law is not a “medical setting”.

Exactly. As you say, the use of the word "addiction" in this context has nothing to do with scientific or medical reality. It's just an ignorant meme of misclassification; like people saying "the internet isn't working" just because their monitor is unplugged. Imagine legislation to "protect the internet" and require by law that all monitors remain plugged in at all times. That is how stupid this reads. The internet working may just mean "monitor displays browser" in common usage, but it doesn't actually mean that and making law based on popular usage is very, very bad.

> As you say, the use of the word "addiction" in this context has nothing to do with scientific or medical reality.

I didn’t say anything like that. I said they’re perhaps using the word in a way that is broader than it’s used in medical jargon. That’s not the same thing as saying the word is meaningless or has nothing to do with reality.

> this law is not a “medical setting”.

And from other posts it's clear that you and those behind this legislation care nothing for science of medical standards when drafting law.

then tell me, why do you need scientific approval before consumption of novel substances, while such other forms of consumption (that access the brain's networks through other modalities than ingestion) are by default "ok"? or are behavioural interventions must be considered as strictly less effective and not require any approval, unless used in a medical setting? and that brings to another question - if it's only the medical setting that decides whether a substance should be considered for approval for consumption (although there's also zero proof that the applications in question thave no psychoactive effects - its rests entirely on your personal feelings), are all other settings simply out of bounds for legal (and, most importantly, moral) considerations? And the way you objectivize such items like "intrinsic value" (completely individual, i.e., case-by-case domain, obviously involved in choice of substance of addiction in all cases of addiction), for me, renders your arguments completely awry. And comparison with conversion therapy is just plain edgy and intentful misrepresentation for the sake of ackchyually.

In my view, this is a perfect case for "there's legitimate doubts about the safety of this technology, so let's ban this until it's completely proven to be safe to consume".

Official release: https://www.governor.ny.gov/news/governor-hochul-joins-attor...

"NetChoice vice president and general counsel Carl Szabo said in a statement that the law would "increase children's exposure to harmful content by requiring websites to order feeds chronologically, prioritizing recent posts about sensitive topics.""

In case anyone was wondering why feeds are not ordered chronologically. Has nothing to do with ads.

At a glance this sounds like fast food.

Are you talking about how they are handling the thing, or the thing itself? If the thing itself, I'd say it is closer to niccotine/bigTobacco more than fast food. These people are intentionally tweaking the algo to make it more addictive just like bigTobacco did.

How can they enforce it?

NY is a huge market both for employment and for ads and sales.

No feeds from addictive algorithms? No notifications after midnight? Can we please raise the age from 18 to maybe 65?

should be for all regardless of age; long as "age" is a required input, there will be incentive to track users one way or another. if its ageless, they have no right to ask for that information (unless they find some other means - which is what we should care about more: stop collecting + tracking users, and parents should be supervising their own kids)

Restricted from to 65, and then also anyone on Medicare, Tricare, or any federally provided medical insurance....

first airbnb, and now this, you love to see it. hopefully banning all gambling is next

I’m always amazed by how much some people want the government to control.

It's only a natural reaction when nothing else is in control. Unless you think that social media platforms are fine as is.

I understand the hesitancy to government control, but at some point, what other mechanism is there? I'm seriously asking.

They weren't referring to Social Media Platforms, they were referring to the previous poster talking about AirBnB and Gambling specifically.

There are many mechanisms we can rely on, like teaching someone the tradeoffs of a dangerous decision.

You say that like addiction is just something that can be taught by parents to overcome. You can attempt to teach your kids to "just say now", but they are kids and are going to rebel against anything they are told they can't do. That's just part of being a kid. Once they do it and become addicted, telling them no is just a useless concept.

Part of the problem of laying it at the parent's feet is that there's a much larger than zero chance the parent(s) also shows addictive behavior. To quote the after school special, "I learned it from watching you, all right?" I still go back to the fact that the addictive thing is being intentionally manipulated to increase the addictive properties. It is not natural. It is the thing in and of itself that is the problem.

I don't know of any viable rehab potential for social media. I can only imagine the relapse potential would be closer to 100% than for any other drug.

I don't intend for my one example to fit all issues. Just trying to point out there is more we can do than rely on government legislation.

The other mechanism is parenting and self-control. I'm not too concerned in the long run about which path is chosen for addictive social media; its days are numbered either way.

This puts parents back in control of what their kids are consuming. Without laws like this, it’s extremely difficult for parents to supervise and control their children’s media consumption.

I get that in theory a kid could sneak off to use TikTok/Instagram on some friend's device away from home, but that's not what's happening. Parents are handing their kids smartphones with unrestricted access to these apps.

People here keep coming back to the social pressures of not allowing their kids on these platforms. It's one thing to give them a device for being able to keep tabs with locations and being able to get in touch with them, but once they have a device where other parents have allowed their kids on the platforms your kid will be made fun of for not being on there.

The parental controls the platforms put on there are non-existent or a mere joke at best (as the recent post about TikTok suggesting porn to minors illustrates). So a parent trying to do things then gets accused of "damaging" their kid by being over restrictive. So, again, what's a parent to do?

A kid won't be ostracized for not having a TikTok. I survived not being on Facebook when it was big in middle school, despite a relatively few dull kids making fun. The same will apply to drugs or gambling by high school, so best learn to deal with it. There are lots of good ways to be popular. Unless of course it's a really bad area, in which case your kid will get beaten up for almost no reason anyway.

But plenty of parents are addicted to TikTok as well. In which case, wouldn't surprise me if they think the kid needs it too.

And I'm always amazed by how much some other people think the free market can do no wrong and should be treated in a completely laissez-faire manner.

If the choice is between the government and the landlord, I choose the government.

I'll choose the one that can't come after me with guns

I'll choose the one that I can at least vote out.

I'll choose the one I can get rid of at will without even having to vote

If you've found a way to participate in society without being constantly surveilled by private companies who frequently leak your most personal information, I'm all ears.

Everyone's pro-democracy until people go "hey maybe allowing a few mega corporations to profit off of addicting everyone to dopamine feeds is actually bad for society."

The biggest con economists ever pulled was to model people as "rational actors" and then generate an entire worldview off of "if everyone acts in their rational self-interest then this works great."

People are almost fundamentally irrational. People are guided by emotion and superstition and a bunch of weird shit that makes it easy for them to get addicted to things that have little to no societal value. If not a government, what force should counter balance that?

All gambling for children? Or?

They should ban never-ending scrolling. User interfaces should be required to divide content into discrete pages.

We need to think bigger. Having an app on your phone is like having a stranger in your house. The same legal protections that apply to property should apply inside the software that runs on your phone. An app should be at the mercy of the user and should provide easy (and automated) ways to turn off manipulative and surveillance features.

An app is at the mercy of the user, who can uninstall at any time, just as you can kick out people out of your home, but you can't turn off "manipulative" behaviour for them.

I can let in a contractor in my house and also ask them not to smoke. They of course have the right to refuse that request and decline my money, but in the real world, contractors usually comply with such requests as smoking is known to harm health. I don't see why the digital world has to be any different. Just because laws and awareness are lagging behind, tech world's toxic practices don't need to get a pass.

> They of course have the right to refuse that request and decline my money, but in the real world, contractors usually comply with such requests as smoking is known to harm health.

Right, but you're suggesting that they should legally have to comply with this request and can't refuse service.

It's the other way around with apps. It's like I want to hire a contractor and every contractor says they have to smoke inside my house.

Which would be legal too, assuming it's a place where smoking is legal (some US cities have banned it in all multi-tenant buildings)

The same legal protections already apply.

You can invite a person into your house to perform a service for you. They can define conditions on performing their service: "I'll shampoo your carpet for $50." Those conditions could also be "I'll shampoo your carpet every month if you let me read your credit card bill every month." You don't have to agree to this! If you don't want to let them read your credit card bill, you don't have to agree to this service.

If you let someone into your house to shampoo your carpet, without agreeing to let them read your credit card bill, and they secretly do that anyway, that's already illegal!

What you're asking is the equivalent of saying that, if someone has a business of shampooing carpets in exchange for reading people's credit card bills, you want to be legally entitled to invite them into your house and force them to shampoo your carpet anyway, without giving them what they want in exchange.

(incidentally - if you respond to my post by nitpicking details of the analogy instead of addressing the central point, I'm not going to bother to respond).

The tech world does not give you the choice of letting you pay for privacy or pay by invading your privacy. Now, you can nitpick that you can build your own service or host your service, but then the question becomes why is programming/sysadmin knowledge a prerequisite for using your phone?

“Here’s a hypothetical ideal analogy, doesn’t that seem reasonable?” is one of my least-favorite types of arguments in cases where the observable reality is extremely different.

Who cares what it might be like? We can see what it is like.

So? You have the choice of not using the service. Companies are not obligated to give you a choice of payment type.

This is pure fiction though. How many people read an app's ToS? If a carpet cleaner showed up with terms like: "I'll shampoo your carpet every month if you let me read your credit card bill every month." no one would read it before signing either, but I suspect that a court would also consider this contract unenforceable.

In addition, with software "contracts", it's often a case of "give an inch and they'll take the mile". The terms are always open to unilateral change from the vendor. So it's more like: "I'll shampoo your carpet for $50. And I can change the terms to whatever I like at any point in the future." which in itself seems insane.

Reply on news.ycombinator.com