5 ms·

> His immediate concern is that the internet will be flooded with false photos, videos and text, and the average person will “not be able to know what is true a

> His immediate concern is that the internet will be flooded with false photos, videos and text, and the average person will “not be able to know what is true anymore.” Isn't this the case already? I expect every post I see in large social media sites posted by somebody I don't personally know to be non-organic feedback by a social media expert. People are doomsaying over a scenario that's identical to the present world.

felipeerias 3y ago
The difference is a matter of scale. In the not-too-distant future, the digital output of LLMs could dwarf the output of humans while being basically indistinguishable from it. At that point, social media will probably split into hyper-local services for people who know each other personally, and an enormous amount of AI-powered rabbit holes for unwary (or depressed, lonely, etc.) users to fall into.
- Double_a_92 3y ago
  So like private group chats and discord servers, and the rest of the internet? ._.
kevincox 3y ago
Yeah, it seems that driving this fact home may in fact be beneficial. Right now a lot of people still assume that everyone on the internet is truthful with good intentions. Making it very clear that this isn't true may be helpful to reset this frame of mind.
- newswasboring 3y ago
  Does anyone else hate the fact that we are actively encouraging making a low trust society? That can't lead to good things...
  - kevincox 3y ago
    Yes, that isn't lost on me but it seems somewhat unavoidable. I would love it if we could trust everyone we meet, but unless we can somehow get everyone to stop doing bad things then you will need to be skeptical of people that you don't have a reason to trust. I think it is important that we add strong mechanisms for trust where possible such as regulations for companies and trust via relationships but it seems that no trust is the inevitable default. People have been running phone scams for decades and it doesn't seem like stopping them from making these attempts is feasible. The most viable solution seems to be helping people identify who they are talking with so that they can avoid falling for the scam. Maybe I am too pessimistic and we will be able to fulfill everyone's needs one day. But as the rich keep getting richer and focusing on power for themselves that doesn't seem like a direction that we are going to move towards soon.
    - newswasboring 3y ago
      First, let me acknowledge I was not trying to single you out. Your comment just prompted that thought. I have seen such talk in a lot of places (and I consume a lot of GPT stuff these days). Sorry if it came off that way. Second, I think you are being too pessimistic. In the sense that you feel like it isn't a solvable problem. I feel like people have thrown their hands up on this problem because in the history of the internet we have not been able to solve it. For me that just looks like a conclusion drawn from insufficient data. To me modern internet is quite young in the grand scheme of things. We haven't even had a generation live out their full life all online. I think we at least should have a hopeful tone about this problem and encourage people to work on it. Presently it feels like everyone is in doom mode.
      - kevincox 3y ago
        I don't feel singled out, it is an interesting train of thought that I also ponder every once and a while. I don't feel like this is specific to the internet. People have been scamming other people for thousands of years. Sure, the internet makes it easier to extend a scammers reach and easier to get away but the fundamental concept is unchanged. Maybe it is a solvable problem, that would be great. But it also seems like this is unlikely to change soon, so at least in the meantime we need to remind ourselves to be careful who we trust.
baby 3y ago
I think this quote is exactly what people are afraid of with the advances of ML, and I think anyone with a bit of mileage browsing the web should be scared as well. It’s a legitimate downside of the tech. It’ll reach a point where you won’t know if the picture you’re looking at, or the voice you’re listening to, or the book you’re reading, or the video you’re watching, is real or generated by AI.

felipeerias 3y ago

The difference is a matter of scale. In the not-too-distant future, the digital output of LLMs could dwarf the output of humans while being basically indistinguishable from it. At that point, social media will probably split into hyper-local services for people who know each other personally, and an enormous amount of AI-powered rabbit holes for unwary (or depressed, lonely, etc.) users to fall into.

Double_a_92 3y ago
So like private group chats and discord servers, and the rest of the internet? ._.

kevincox 3y ago

Yeah, it seems that driving this fact home may in fact be beneficial. Right now a lot of people still assume that everyone on the internet is truthful with good intentions. Making it very clear that this isn't true may be helpful to reset this frame of mind.

newswasboring 3y ago
Does anyone else hate the fact that we are actively encouraging making a low trust society? That can't lead to good things...
- kevincox 3y ago
  Yes, that isn't lost on me but it seems somewhat unavoidable. I would love it if we could trust everyone we meet, but unless we can somehow get everyone to stop doing bad things then you will need to be skeptical of people that you don't have a reason to trust. I think it is important that we add strong mechanisms for trust where possible such as regulations for companies and trust via relationships but it seems that no trust is the inevitable default. People have been running phone scams for decades and it doesn't seem like stopping them from making these attempts is feasible. The most viable solution seems to be helping people identify who they are talking with so that they can avoid falling for the scam. Maybe I am too pessimistic and we will be able to fulfill everyone's needs one day. But as the rich keep getting richer and focusing on power for themselves that doesn't seem like a direction that we are going to move towards soon.
  - newswasboring 3y ago
    First, let me acknowledge I was not trying to single you out. Your comment just prompted that thought. I have seen such talk in a lot of places (and I consume a lot of GPT stuff these days). Sorry if it came off that way. Second, I think you are being too pessimistic. In the sense that you feel like it isn't a solvable problem. I feel like people have thrown their hands up on this problem because in the history of the internet we have not been able to solve it. For me that just looks like a conclusion drawn from insufficient data. To me modern internet is quite young in the grand scheme of things. We haven't even had a generation live out their full life all online. I think we at least should have a hopeful tone about this problem and encourage people to work on it. Presently it feels like everyone is in doom mode.
    - kevincox 3y ago
      I don't feel singled out, it is an interesting train of thought that I also ponder every once and a while. I don't feel like this is specific to the internet. People have been scamming other people for thousands of years. Sure, the internet makes it easier to extend a scammers reach and easier to get away but the fundamental concept is unchanged. Maybe it is a solvable problem, that would be great. But it also seems like this is unlikely to change soon, so at least in the meantime we need to remind ourselves to be careful who we trust.

baby 3y ago

I think this quote is exactly what people are afraid of with the advances of ML, and I think anyone with a bit of mileage browsing the web should be scared as well. It’s a legitimate downside of the tech. It’ll reach a point where you won’t know if the picture you’re looking at, or the voice you’re listening to, or the book you’re reading, or the video you’re watching, is real or generated by AI.