Hacker News Reader

Comment by ethbr1 | Hacker News Reader

Thought experiment -- should it be illegal to provide an AI chatbot to children that indoctrinates them with religious beliefs?

Because we're edging closer to that too: https://www.scientificamerican.com/article/the-god-chatbots-...

View on HN

Chatbots and social media should be banned for 16 YOs and under.

Completely agree. I believe that social media overstimulates a young person’s expectations of society in the same way that porn overstimulates our expectations of sex.

Dunno, I feel porn gave me quite positive and reasonable expectations of sex (relaxed, that it is fun, that women like sex too, etc). It made sex seem much less dramatic and more normal. But maybe I am an outlier plus it is not like I started watching porn until I was like 17-18.

I am sufficiently old that I did not experience hard core internet porn until I could manage it. But evidence seems to show that for the vulnerable, porn consumption can lead to dopamine deletion and depression.

> It made sex seem much less dramatic and more normal.

I think you’ll find quite a lot of people, even here, view this as a bad thing.

Probably also depends on the porn you were exposed to. I think the people who see porn as inherently problematic are usually mostly thinking of the nastier stuff.

Where is the line and what defines the boundaries for what is termed social media?

For example, is discord still permitted? Ai bots and bad influences exist there as well, but I don't think a ban is the right solution.

> Where is the line and what defines the boundaries for what is termed social media?

Algorithmic sorting + public or semi-public content. Chat rooms have different problems.

> Algorithmic sorting + public or semi-public content.

That includes HN, among other things.

Putting age limits on sites requires age verification for everyone. And no, there isn’t a clever crypto mechanism that makes anonymous age verification work without also making it easy for people to borrow age verification credentials from someone older.

> That includes HN

I think when people say “algorithmic sorting” they usually mean an algorithm which generates different, “personalized” sort order for different users.

Australia basically just came up with a list of social media sites it applies to. It's perhaps not entirely fair or objectively true. But it lets them age restrict Facebook/Tiktok/etc without restricting IMs or small sites like HN.

From my experience as a teacher, I believe that ticktock and instagram are the worst offenders, particularly for young women. The hyper-visuality and ease of consumption of these media sets them apart from platforms which can accommodate actual discussion (such as discord). The very fact that ‘influencer’ is now a profession supports my position.

That being said, I am not of the ‘for gods sake won’t someone think of the children’ brigade. Their goal seems to be to use the vulnerability of young people to control the internet.

Also, the emphasis on final result, without accurately portraying the work that went into it.

It's probably not healthy for younger people to be able to swipe through the finished products of 40+ hours of work, which the videos make seem like just happened.

Australia has just banned "social media" for under 16s.

As someone who got a lot of positive value out of parts of social media from around age 14, I think this needs to be done in a more careful way than it was done here. Specifically, I don't think that communication apps such as WhatsApp/Messenger/etc should be banned as they form a key part of communication in and out of school, staying in touch with family, etc.

What I'd like to see is more nuanced laws around the exposure of children to social media algorithms. For example, should 14 year olds be on Instagram? Well Instagram DMs are the chat app of choice for many people, so they should probably get that (with safety and moderation controls). How about the public feed? Maybe? But maybe there shouldn't be likes/dislikes/etc. Or maybe there shouldn't be comments.

Without getting into the weeds over whether they should have done this at all, thought has been given to exactly the issues you just raised:

    The laws, which will come into effect from late 2025, will bar under-16s from being able to access social media platforms such as Facebook, Instagram, Snapchat, Reddit and X.

    Exemptions will apply for health and education services including YouTube, Messenger Kids, WhatsApp, Kids Helpline and Google Classroom

~ https://www.theguardian.com/media/2024/nov/29/how-australias...

It's important that there's a means of communication between parents and kids, it doesn't have to be Instagram DM's and if that's no longer available the history of the internet to date suggests that habits would change and switch to whatever is available.

The Aussie law does allow WhatsApp and Messenger Kids, among others. I agree we need nuance for these types of laws. We also need the realistic acknowledgement that kids are usually more savvy than their parents and any law that is too strict will just drive kids to find alternatives that have less transparency, less moderation and less accountability.

And though I know the age limits on these things are necessarily arbitrary, I do wish we would accept that 16-year-olds are not kids. Many of them are driving, working part-time jobs, having sex and relationships, experimenting with drugs, engaging with religion and philosophy, caring about politics... the list goes on. They may not be adults, but if we have to have an arbitrary cutoff for the walled-off "kids world" we want to pretend we can create, it can't extend all the way to 16-year-olds.

It seems healthy to put a financial restraint on organized social media companies profiting off pre-adults.

Less profit motive = less attention hacking = more room for user-positive engineering

> The Aussie law does allow WhatsApp and Messenger Kids

...but not Instagram DMs I don't think, which is functionally a chat app and used in that way.

Is instagram DMs a separate app that can be used without interacting or signing up for the social media part?

I don't think we know wtf they've actually banned until we see the implementation details.

It could be a Pornhub-style "are you over 16?" checkbox, all the way up to building a Chinese-style identity verification system.

The worst case is that VPN use in Straya will rise a lot next year, and all the ban will effectively have done is removed parental oversight on the kids' social media accounts.

edit: spelling, and also rethinking "the worst case". The worst case is some idiotic uniquely-Australian authoritarian attempt to remove online anonymity that ends up making a complete NBN-style farce of the place.

I think no-one should be exposed to social media and doom scrolling.

Still it shouldn't be banned away, it's the parents responsibility to control access for their kids and everyone should be smart enough to avoid social media.

If you don't, that's on you, suit yourself.

The article mentions 17 yos.

Then while we're at it, move that to 18 so it's only for responsible adults.

Then let's also forbid people with known mental issues. Or really there should be a check for each ser to see how much they can handle psychotic responses.

Just punting it to "user's should be older" is the lazy answer IMHO, if we're going to limit let's put actually legit barriers.

Let's think of the converse: should it be illegal to provide an AI chatbot to children that indoctrinates them with anti-religious beliefs (i.e., it talks them out of religious beliefs)? And what if this conflicts with the child's indoctrination in religious beliefs by those parents? And what if those religious beliefs are actively harmful, as many religious beliefs are? (see any cult for example)

To turn this around, why are parents allowed to indoctrinate their children into cults? And why is it a problem if AI chatbots indoctrinate them differently? Why is it held as sacrosanct that parents should be able to indoctrinate children with harmful beliefs?

Thanks for link, interesting reading.

I don't know whether to be horrified (because you're right, the implication for teenage religious radicalisation is horrifying), or quietly pleased that LLMs will reduce the political power of churches.

Given that it’s not illegal to send children to schools that indoctrinate them with religious beliefs I’d speculate that the answer is a resounding “no”.

Reply on news.ycombinator.com