• 21 Posts
  • 274 Comments
Joined 1 year ago
cake
Cake day: August 28th, 2023

help-circle

  • Problem: Higher childhood depression rates linked to social media usage, social media caused disruption in education (like usage in schools), privacy violation of minors, etc.

    An enforceable, common sense solution: Very strict privacy protection laws, that would end up protecting everybody, including minors. Better, kid friendly urban infrastructure like dedicated bike paths protected from car traffic, better pedestrian areas, parks and so on. Kids will get outside their house if there is a kid friendly outside. A greener, more human friendly outside where you can socialize with other humans would always be preferred over doom scrolling online. For the disruption in education issue, it is very education system dependent.

    What solution these people came up with: Make it illegal for individuals under the age of 16 to create social media accounts. How do they enforce this? No idea. Does this solve any of the above problems? No. Is this performative? Yes.

    Speaking from personal experience, social media was one of the most liberating tools for me as a kid. I lived in a shitty, conservative country and was gay. Social media told me that I wasn’t disgusting. I was always more of a lurker than a poster, so I thankfully didn’t really experience being contacted by groomers and so on. However, many of my friends who posted their images and stuff almost always got pedos in their DMs, so that’s a very real issue.

    I could ask my silly little questions related to astrophysics on Reddit and get really good answers. Noone around me irl was ever interested/able to talk about stuff like this. I could explore different political ideologies, get into related servers on Discord and learn more about this. None of this was possible without social media.

    Banning social media outright is such a boomer move lol. Doing so isn’t going to solve any real problems associated with childhood social media usage. It’s just going to give the jackass parents complaining about this a false sense of security, when the kids still end up suffering.




    1. You are assuming that the current medical scene won’t improve. It is very likely that we’ll eliminate the “old person lying in bed, dying” visual altogether due advancements in the medical field (especially accelerated further by development in AI)
    2. The “human touch” is not impossible to replicate for machines. You aren’t seeing machines capable of that right now, because the field of personal care robots are in their absolute infancy. “The human touch” at the end of the day, is just warm, soft skin paired with a caring voice. We have already replicated the caring voice.
    3. Elder care robots won’t be cold, metal bodies going “Boop boop, shit in bed defected, Boop boop engaging cleanup procedure…”. They would be really kind voices, soft hands with an experience of more than a thousand years of handling thousands of patients. They would never become impatient, they would never feel bad or disgusted.

    Of course, advancements in this tech won’t stop humans from caring for the elderly. You can still care for ur grandpa. However, ur grandpa won’t die if u don’t.

    Here’s the best case scenario - you can be with ur grandpa, chat, play video games, do fun stuff. When it’s time to change the diaper, a professional robot trained for this very purpose does the job.



  • Exactly. There’s a very clear path to monetisation for the bigger tech companies (ofc, not the random startup that screams “AI quantum computing blockchain reeeee”).

    Lemmy is just incredibly biased against AI, as it could replace a shit ton of jobs and lead to a crazy amount of wealth inequality. However, people need to remember that the problem isn’t the tech- it’s the system that the tech is being innovated in.

    Denying AI is just going to make this issue a lot worse. We need to work to make AI be beneficial for all of us instead of the capitalists. But somehow leftist talk surrounding AI has just been about hating on it/ denying it, instead of preparing for a world in which it would be critical infrastructure very soon.


  • I have no idea how people can consider this to be a hype bubble especially after the o3 release. It smashed the ARC AGI benchmark on the performance front. It ranks as the 175th best competitive coder in the world on Codeforces’ leaderboard.

    o3 proved that it is possible to have at least an expert AGI if not a Virtuoso AGI (according to Deep mind’s definition of AGI). Sure, it’s not economical yet. But it will get there very soon (just like how the earlier GPTs were a lot dumber and took a lot more energy than the newer, smaller parameter models).

    Please remember - fight to seize the means of production. Do not fight the means of production themselves.






  • Not we. It was Jeffery Epstein. Aliens wanted to meet the smartest human. So they came down to Epstein Island to meet Stephen Hawking. Jeffery mistook them for kids (the Greys are short, everyone knows that).

    After Jeffery got to know that the “kids” he raped were actually aliens, he was very apologetic. He tried to explain the issue, but the aliens were too traumatised. They went back to Kepler-22b for a few months.

    After Jeffery got put in prison, the aliens came back for revenge. They used a mind control device to make Jeff hang himself.

    So in conclusion, Jeff was not killed by his rich pervert friends (oh god, why would you even say that). It was the aliens (no, not the “illegal transgender aliens” Trump keeps talking about)


  • Hm, makes sense ig. Basically, what u’r saying is this from what I understood - AI romance/sex bots capable of making a significant drop in birth rates would come before AI bots that bring in labor post scarcity.

    While I agree with this, I don’t think that the time difference between the two events would be significant enough for the drop in birth rate to be that damaging. Why? Because I’m assuming that development in AI would be that fast. I can’t think of many reasons as to why tech that makes it possible to serve as a good enough romantic partner (which is quite a complex task) can’t serve as a mental health therapist (with different fine tuning of course), customer service, retail, admin, secretary, etc.

    One doesn’t need to replace 100% of jobs to cause unemployment related issues in the market. I think the effects of unemployment would be seen first before the effects of potentially dropping birth rates.


  • Your conclusion is based upon an assumption that we need more humans to progress as society. If AI develops to the point where it is better as a partner than a human being, it likely means that we have achieved, or are very close to achieving labor post scarcity (the assumption being that an AI capable of achieving this is capable enough to do most/all human work).

    When we achieve labor post scarcity, the number of humans has nothing to do with progress. Therefore, falling birthrates won’t have any negative effect on progress.

    When we achieve labor post scarcity in the medical field, life expectancy would increase, with us achieving biological immortality at a certain point. This means, that death rates also go down.

    Considering the above, I thought you were referring to “dating and fucking AI partners” as the end of human progress (presumably because of a lack of any motivation to cause any more development). That’s what my reply was talking about.




  • It seems that LIDAR based self driving tech (the one that Waymo uses) is miles ahead than 2d camera based tech (the one that Tesla uses).

    I’m really excited for this being implemented for city buses. Currently, in lower density areas, there are two choices - have smaller buses with more frequency, or have larger buses with less frequency. The problem with smaller buses is that you need more drivers. The problem with larger buses is that frequency becomes low, thus disincentivizing usage of public transit.

    Self driving city buses would be really cool as driver costs would be 0. This would mean, that smaller buses/vans would be able to run at fixed routes at really high frequencies.