Georgios (er/ihn)
JA WestenbergNo, You Shouldn't Let Your Kids Use ChatGPT. A thread. đ§ľ
1/16
We pretend that because the interface is clean and thereâs no nicotine, no violence, no nudity, that itâs safe. It looks like a homework helper. A science fair assistant. A miracle of modern education.
Thatâs just marketing.
2/16
You wouldnât let your child hang out unsupervised with a stranger - especially one who lies confidently, speaks with artificial authority, and occasionally invents facts.
3/16
But thatâs what weâre doing when we let them talk to generative AI with no guardrails and no context. It looks smart. It feels friendly. It sounds right. Thatâs exactly what makes it dangerous.
4/16
We underestimate how deeply plastic the young mind is.
Kids donât use tools; they internalize them.
5/16
Kids learn how to think by watching thinking happen. When you train on a language model, it doesnât learn truth, it learns patterns. When a kid trains on a language model, the same thing happens. They start seeing speech as performance.
6/16
They start believing fluency equals wisdom. They mimic the mimicry.
7/16
We donât give a five-year-old a credit card and say, âGood luck budgeting.â We donât drop a 10-year-old into Times Square at midnight and call it a field trip.
8/16
We create buffers. We wait until theyâve got context, maturity, the ability to weigh signal from noise.
And even then, we supervise.
9/16
ChatGPT etc are powerful - and fundamentally misaligned with how kids learn to trust, reason, and discern.
10/16
These models shape the questions you ask next. They donât reflect your thinking. They nudge it. Relentlessly.
11/16
I'm not trying to create a panic. This is a boundary. If you wouldnât let your kid join Twitter, if you wouldnât let them Google health symptoms unsupervised, donât let them outsource cognition to a system you donât understand.
12/16
Curiosity needs friction. Learning needs surprise. Wisdom needs mistakes. Models donât offer that. They offer something faster, smoother, and emptier.
13/16
We can teach kids to use these tools with judgment, with context, with skepticism. But that starts with a pause. With an adult in the room. With a conversation about what these models are and what theyâre not. It starts with treating intelligence as more than output.
14/16
Once you flatten knowledge into prediction, once you replace the actual road of learning with a shortcut that feels smarter than you are, youâve done more harm than you know.
15/16
Youâve reshaped the map your kid is using to navigate the world.
Youâve said: hereâs something that sounds like thinking.
Something easier than thinking.
Good luck un-ringing that bell.
16/16
Hartmut Seichter@Daojoan - exactly đ - we ignore the fact that it is in the vested interest of these tech companies that we get addicted to these dethinking machines. Once we are hooked on them, we lack the means of critical thought to question what the purpose is and what we are as human beings
Nicole ParsonsTake the AI hype in context:
1. Joni Ernst "all going to die anyway" nihilism
2. Elon Musk's "empathy is for the weak" narratives.
Habituating your child to treating people like they treat an AI device.
Frank Herbert -- Dune
"The devices themselves condition the users to employ each other the way they employ machines."
@Npars01 only just catching up on the Ernst stuff. Wow.
Kid Mania@Daojoan @Npars01
Do NOT let your children anywhere near LLMs (so-called "AI")...
Here are a couple of reasons:
https://mastodon.social/@ekis/114613560446815567
and even more disturbing:
https://mastodon.social/@ekis/114613739460407851
@Daojoan @Npars01
And if that's not good enough, how about this...?
People have literally started to worship it. #doomed
"People Are Losing Loved Ones to AI-Fueled Spiritual Fantasies
Self-styled prophets are claiming they have 'awakened' chatbots and accessed the secrets of the universe through ChatGPT"
https://archive.ph/eeesj#selection-1485.0-1485.61
P_______X
Little Art Histories@Npars01 @retrakker @Daojoan Thank you, I was thinking forever of this quote but couldn't figure out where it came from.
Dr Andrew A. Adams #FBPE đś@retrakker @Daojoan
See also a simple nudged AI being very effective in pushing people in the way they're being nudged.
https://researchmap.jp/acubed/published_papers/25006202
See also a simple nudged AI being very effective in pushing people in the way they're being nudged.
https://researchmap.jp/acubed/published_papers/25006202
ă˘ăă ăš ă˘ăłăăŞăĽăź (Andrew A Adams) - The Effects of Nudging a Privacy Setting Suggestion Algorithm's Outputs on User Acceptability - čŤć - researchmap
researchmapăŻăćĽćŹăŽç 犜č ć ĺ ąăĺéăťĺ Źéăăă¨ă¨ăăŤăç 犜č çăŤăăć ĺ ąçşäżĄăŽĺ ´ăç 犜č çăŽéăŽć ĺ ąäş¤ćăŽĺ ´ăćäžăăăă¨ăçŽçă¨ăăŚăĺ˝çŤç 犜éçşćłäşşç§ĺŚćčĄćŻčćŠć§ďźJSTďźăéĺśăăăľăźăăšă§ăă
@retrakker @Daojoan
See also a simple nudged AI being very effective in pushing people in the way they're being nudged.
https://researchmap.jp/acubed/published_papers/25006202
See also a simple nudged AI being very effective in pushing people in the way they're being nudged.
https://researchmap.jp/acubed/published_papers/25006202
ă˘ăă ăš ă˘ăłăăŞăĽăź (Andrew A Adams) - The Effects of Nudging a Privacy Setting Suggestion Algorithm's Outputs on User Acceptability - čŤć - researchmap
researchmapăŻăćĽćŹăŽç 犜č ć ĺ ąăĺéăťĺ Źéăăă¨ă¨ăăŤăç 犜č çăŤăăć ĺ ąçşäżĄăŽĺ ´ăç 犜č çăŽéăŽć ĺ ąäş¤ćăŽĺ ´ăćäžăăăă¨ăçŽçă¨ăăŚăĺ˝çŤç 犜éçşćłäşşç§ĺŚćčĄćŻčćŠć§ďźJSTďźăéĺśăăăľăźăăšă§ăă
mozziediver@Daojoan
Excellent!
Here it is in an ordered format in MastoReader.
https://mastoreader.io/?url=threadhttps%3A%2F%2Fmastodon.social%2F%40Daojoan%2F114612369770680024
Excellent!
Here it is in an ordered format in MastoReader.
https://mastoreader.io/?url=threadhttps%3A%2F%2Fmastodon.social%2F%40Daojoan%2F114612369770680024
tjal@Daojoan first hand experience of this on reddit agrees with this: https://www.reddit.com/r/ChatGPT/s/OSIbnXh6oA
Chris@Daojoan I'm pretty sure this danger exists (to a lesser extent maybe, or it takes longer) for adults as well.
Tordenstorm@thechris @Daojoan Strongly agree! I don't want to sound tyrannical but I told my significant other to avoid ChatGPT, since it's the easiest way to become brainwashed and misinformed. (I feel myself an abusive, control-freak partner for not allowing a so-called AI to abuse and control my love. Quite ironical.)
Erik Jonker@Daojoan great thread, additional complexity, it's not about "ChatGPT" only anymore, the technology is integrated in Google Search and many other applications, visible or less visible. If we want to limit kids, limit screentime is the only way i guess but how realistic is that with current smartphone use...also national government and EU are slow or not regulating enough, a hard time to raise kids (mine are 18+ so i just coach them with regard to AI they use regularly by the way...)
@ErikJonker All true. And - as a parent - these are the things keeping me up at night.
Cassandrich@ErikJonker @Daojoan As always, the kids are the savvy ones who know this shit is cringe garbage. It's the adults who are a threat. But adults want to take away kids' access to friendships and information as punishment for adults being awful... đ đ¤Ź
Serg Metelin@dalias @ErikJonker @Daojoan I wish kids would be that self-aware, but sadly that's not what happened to kids of 2010s when iPhones with selfie cameras and Instagram came out... Broken generation of kids :(
@serg @ErikJonker @Daojoan "Broken generations of kids" is something only members of broken generations of olds would say or even think.
@dalias @ErikJonker @Daojoan I wish you were right, but that's not what the data shows. Check out https://app.thestorygraph.com/books/0ae97ada-0a45-478a-a11a-ec8d01d688d7 and maybe it will change your mind.
To draw a parallel - do you think it's ok to market and sell cigarettes to kids? Do they have enough insight to know that it's cringe?
@dalias @ErikJonker @Daojoan also, we all are broken atm - kids and adults. I guess we all are trying to figure out how to get out of this shitty trap we got ourselves into.
@Daojoan I feel like almost all this applies equally if you replace "generative AI" by "social media"
@vegard I donât disagree with that
@Daojoan đI think there is also one other issue. Those agents are super enthusiastic, always polite and happy to explain. It is a pleasant experience to chat about some topic of interest. Our human relationships are not like that - we have bad days, sometimes loose patience, get bored with a topic.
As we grow up we learn how to handle such situations. I'm concerned people may start expecting others around to act like a chatbot and once that fails, loose all motivation to build relationships.
As we grow up we learn how to handle such situations. I'm concerned people may start expecting others around to act like a chatbot and once that fails, loose all motivation to build relationships.
Lappenjammer@Daojoan Plenty of food for thought in your thread. I don't even have kids and benefit from reading it. Thanks!
MillardPhillmore@Daojoan đŻ concurrently, this is happening to "adults"... I'm witnessing it at my place of work in real time.
Photo Ewen đˇWe are all of us a product of our environment. Cannot imagine what kind of dysfunction might result for a child growing up thinking their ChatGPT sessions are legitimate sources for role modelling.
It's like what Fox News did to our grandparents.
Extinction the Sixth
sb arms & legs@Daojoan
The men in charge of these llm megaprojects are ivy league educated, lifelong readers of the science fiction of the last century - they read the stories predicting exactly what we see happening, and exactly what is being explained here.
What if that *is* the point? If alongside profit for themselves, the reason LLMs must replace anything they can as fast as they can is to hasten the atrophy of critical thinking, replacing it with a dependance on - who other - big tech oligarchs.
They also read #yarvin.
Mercurius GoldsteinLLMs will some day be remembered, if they are remembered at all, as the lead paint/leaded fuel/DDT of computing.
Mary Beth Rudis@Daojoan important issues you bring up; but arenât you sort of âpreaching to the choirâ? some of us witnessed the failure of 2005-2008 first hand: which was educate the educators first⌠about how companies are monetizing human behavior, how K-12 is being targeted by corporate greed, and once you teach the teacher, they take it from there (turns out they are really smart, just need to be shown); we failed to educate teachers abt social media in its infancy; letâs not fail them now
DistroWatch@Daojoan I see this sometimes with younger kids who are just used to solving problems by looking up the answer or watching a video. They don't have their own trial and error approach so if they can't see a video with the answer they are stuck. It makes them dependent on a source, any source, of answers.
CubeThoughts@Daojoan This is great!
But for adults to have a conversation about what the models are and what they're not, and I feel that adults in general have already been conditioned to equate confidence with knowledge, expertise and wisdom.
Txo!?
ellieAnd wronger.
Examine the motives of AI's investors.
https://www.bloomberg.com/news/newsletters/2025-05-13/trump-mbs-tout-1-trillion-saudi-deal-commitment
https://www.washingtonpost.com/opinions/2025/05/29/mbs-saudi-arabia-big-tech-authoritarian-embrace/
Do you want your child anywhere near this man?
https://www.thenation.com/article/society/wall-street-mbs-saudi-al-rumayyan/
https://www.cnbc.com/2018/04/07/heres-a-look-at-who.html
https://www.businessinsider.com/saudi-arabia-crown-prince-visits-apple-google-2018-4
https://www.nytimes.com/2025/05/30/us/elon-musk-drugs-children-trump.html
Or these ones?
https://www.nytimes.com/2025/05/30/technology/trump-palantir-data-americans.html
https://newrepublic.com/post/195904/trump-palantir-data-americans
https://www.404media.co/larry-ellisons-ai-powered-surveillance-dystopia-is-already-here/
SorceryForEva đłď¸âđ đłď¸ââ§ď¸Worse, we cannot sense this relentless nudging because we do not know what we do not know.
Sci-fi lied to me when it suggested I would be able to sense mind-affecting technologies!
I could hardly imagine a better mechanism than modern "AI" to invisibly guide a mental trajectory from a reasonable, curious starting point to an arbitrarily extreme position.
But sure, accepting this incredible power from corporate hands is perfectly safe. No possibility of ulterior motives there!
Mike Taylor đŚ@Daojoan BTW., this also the outcome of the English Public School system.
(Note that "Public School" in England doesn't mean state-funded school, it means extremely expensive elite top-end private school.)
Fiona@Daojoan Unfortunately "believing fluency equals wisdom" is already well-entrenched in adults, too. That's essentially why LLMs are catching on.
Where I disagree is calling them "powerful", they're worse than useless, unless what you actually want is "making up stuff that sounds plausible at the surface", a propaganda amplifier, if you will.
Where I disagree is calling them "powerful", they're worse than useless, unless what you actually want is "making up stuff that sounds plausible at the surface", a propaganda amplifier, if you will.