4 ms·

Well, the way I see it, we have spent the last 20 years building Social Media, Influencer culture, Creators, and a Ad-tech driven surveillance. If AI helps us

Well, the way I see it, we have spent the last 20 years building Social Media, Influencer culture, Creators, and a Ad-tech driven surveillance. If AI helps us focus on Robotics, Spaceships, and underwater cities, I am game.

dougmwne 4y ago
Yeah, I don’t get the panic. We’re not on any good trajectory here. All signs point to suck. At least we have a chance to pivot.
raincole 4y ago
The most popular AI model right now is LLM. Does LLM help "Social Media, Influencer culture, Creators, and a Ad-tech driven surveillance" or "Robotics, Spaceships, and underwater cities" more?
- IIAOPSW 4y ago
  Robots and shit. What value is social media when its all just a big fat LLM dumping ground and everyone knows it. Long Live the Machine, killer of the bird and the book. That is my prophecy.
- TeMPOraL 4y ago
  That's a bit unclear at this point. Around GPT-3 it became apparent that LLMs are no longer glorified Markov chains - the training sets grew large enough, and the latent space high dimensional enough, that the model starts to capture the kind of fuzzy associations humans make. That is, it seems to capture an important subset of what we consider thinking. So I guess GPT-4 may still be most useful at destroying the modern information economy, thanks to being an infinite generator of garbage indistinguishable from content marketing. But GPT-5 might be just good enough to enable some qualitative breakthroughs in robotics, spaceships and underwater cities... (NOTE: this is not be believing that OpenAI solved consciousness. This is me recognizing the surprising ability for generative language models to navigate abstract concept space, just by virtue of being exposed to enough association implicitly encoded in everything we say or write...)
- jhanschoo 4y ago
  > Social Media, Influencer culture, Creators, and a Ad-tech driven surveillance I think it does help those. You've seen how the early models had to be taken off because teenagers trained models to be bigots. If there's a widely-deployed model that uses near-contemporary data, it would be the prime target for all kinds of promotion, influence, disinformation and product placement just like how Google became one such. The reasonable, useful replies that current models give you may become a thing of a past once people start to target LLMs for the same things that they today target Google's search, and you will have the same SEO arms race, albeit probably harder to combat considering how much more opaque LLMs probably are compared to Google's search algorithms. Perhaps the good useful information that we get from LLMs today are more a product of them being trained on paid content of high quality that aren't widely distributed.
TeMPOraL 4y ago
The irony is: "Social Media, Influencer culture, Creators, and a Ad-tech driven surveillance" are what enabled the AI advance. All that "user generated content" we kept posting on centralized platforms is exactly what got blended and turned into the breakthrough ML models of the past ~5 years. GPT family literally exists because of Reddit. Without the ad-driven Internet forcing the world to generate so much content, garbage or otherwise, I'm not sure if and when we'd have generative models.
- jsemrau 4y ago
  I guess that's where this goes. Just a flood of even more irrelevant content making all Social Media apps inherently worthless.
  - TeMPOraL 4y ago
    Yes, it seems to be a one-off, self-limiting thing. Not only the flood of garbage will destroy the Internet for us, it'll destroy its utility for ML models too. Unless, the current models can effectively spam-filter their own output, which... is a possibility. I wonder if the Internet Archive are aware of the importance their job is about to gain. They're not just maintaining an archive of the Internet - they're maintaining an archive of pre-generative AI Internet, meaning they effectively hold what's about to become the main source of pristine, human-generated content for training corpora.
somewhereoutth 4y ago
The problem with Robotics, Spaceships, and underwater cities is that failure can be catastrophic. Whereas if you get served the wrong ad occasionally, it is fine.
- jsemrau 4y ago
  Well in the 60s we "[...] choose to go to the Moon in this decade and do the other things, not because they are easy, but because they are hard". Now we are afraid of our add showing up in odd places. Of course I am exaggerating, but it shows how little impact ad-placement have on our actual lives.
ChatGTP 4y ago
If AI helps us focus on Robotics, Spaceships, and underwater cities, I am game. I think this is "the promise" but I don't think it will be the outcome, like many things out of Silicon Valley. Remember when Facebooks mission was to "connect the world", wow, what a fail.

dougmwne 4y ago

Yeah, I don’t get the panic. We’re not on any good trajectory here. All signs point to suck. At least we have a chance to pivot.

raincole 4y ago

The most popular AI model right now is LLM. Does LLM help "Social Media, Influencer culture, Creators, and a Ad-tech driven surveillance" or "Robotics, Spaceships, and underwater cities" more?

IIAOPSW 4y ago
Robots and shit. What value is social media when its all just a big fat LLM dumping ground and everyone knows it. Long Live the Machine, killer of the bird and the book. That is my prophecy.
TeMPOraL 4y ago
That's a bit unclear at this point. Around GPT-3 it became apparent that LLMs are no longer glorified Markov chains - the training sets grew large enough, and the latent space high dimensional enough, that the model starts to capture the kind of fuzzy associations humans make. That is, it seems to capture an important subset of what we consider thinking. So I guess GPT-4 may still be most useful at destroying the modern information economy, thanks to being an infinite generator of garbage indistinguishable from content marketing. But GPT-5 might be just good enough to enable some qualitative breakthroughs in robotics, spaceships and underwater cities... (NOTE: this is not be believing that OpenAI solved consciousness. This is me recognizing the surprising ability for generative language models to navigate abstract concept space, just by virtue of being exposed to enough association implicitly encoded in everything we say or write...)
jhanschoo 4y ago
> Social Media, Influencer culture, Creators, and a Ad-tech driven surveillance I think it does help those. You've seen how the early models had to be taken off because teenagers trained models to be bigots. If there's a widely-deployed model that uses near-contemporary data, it would be the prime target for all kinds of promotion, influence, disinformation and product placement just like how Google became one such. The reasonable, useful replies that current models give you may become a thing of a past once people start to target LLMs for the same things that they today target Google's search, and you will have the same SEO arms race, albeit probably harder to combat considering how much more opaque LLMs probably are compared to Google's search algorithms. Perhaps the good useful information that we get from LLMs today are more a product of them being trained on paid content of high quality that aren't widely distributed.

TeMPOraL 4y ago

The irony is: "Social Media, Influencer culture, Creators, and a Ad-tech driven surveillance" are what enabled the AI advance. All that "user generated content" we kept posting on centralized platforms is exactly what got blended and turned into the breakthrough ML models of the past ~5 years. GPT family literally exists because of Reddit. Without the ad-driven Internet forcing the world to generate so much content, garbage or otherwise, I'm not sure if and when we'd have generative models.

jsemrau 4y ago
I guess that's where this goes. Just a flood of even more irrelevant content making all Social Media apps inherently worthless.
- TeMPOraL 4y ago
  Yes, it seems to be a one-off, self-limiting thing. Not only the flood of garbage will destroy the Internet for us, it'll destroy its utility for ML models too. Unless, the current models can effectively spam-filter their own output, which... is a possibility. I wonder if the Internet Archive are aware of the importance their job is about to gain. They're not just maintaining an archive of the Internet - they're maintaining an archive of pre-generative AI Internet, meaning they effectively hold what's about to become the main source of pristine, human-generated content for training corpora.

somewhereoutth 4y ago

The problem with Robotics, Spaceships, and underwater cities is that failure can be catastrophic. Whereas if you get served the wrong ad occasionally, it is fine.

jsemrau 4y ago
Well in the 60s we "[...] choose to go to the Moon in this decade and do the other things, not because they are easy, but because they are hard". Now we are afraid of our add showing up in odd places. Of course I am exaggerating, but it shows how little impact ad-placement have on our actual lives.

ChatGTP 4y ago

If AI helps us focus on Robotics, Spaceships, and underwater cities, I am game. I think this is "the promise" but I don't think it will be the outcome, like many things out of Silicon Valley. Remember when Facebooks mission was to "connect the world", wow, what a fail.