4 ms·

I find that hard to believe. The AI companies will want to control what's possible and find new things to do that "need" their services. Otherwise it would be l

I find that hard to believe. The AI companies will want to control what's possible and find new things to do that "need" their services. Otherwise it would be like Intel and Microsoft had decided in the year 2000 that computers are "good enough" now and we would have explored what's possible with that hardware ever since.

benterix 4mo ago
> The AI companies will want to control what's possible and find new things to do that "need" their services. That's correct. The problem is they have smart people, tons of money, and several years to figure that out, and the best thing they can come up is a coding agent.
- lazide 4mo ago
  That isn’t the best thing they’ve come up with. It’s a marquee product that is fit for public consumption, however. The ‘best’ things are; - fuzzy pattern matching algorithms for traffic analysis, human and other image target recognition. - targeting algorithms that identify ‘suspicious’ individuals in large volumes of metadata. - fraud analysis - antagonistic image and video generation, both for fooling other fraud analysis, but also for propaganda, screwing with other actors, etc. - directed high speed content generation (text, pictures, video) to spam the ‘algorithm’ and allow near realtime identification of additional buttons to push for given target audiences. - massive marketing/ad manipulation. Those budget line items (and the suppliers) really want to stay off the radar however, as it makes their life harder.
  - benterix 4mo ago
    But you're mentioning several things that predate the current LLM craze and belong to the ML domain. These mostly benefit from GPUs but often have much lower hardware requirements. I'm talking specifically about the moat of LLM providers.
    - lazide 4mo ago
      Sure, but all fall under the same marketing umbrella. Ad/marketing manipulation are exceptionally well done with LLMs in particular. If you asked someone if drone auto targeting/image recognition or data analysis was ‘AI’, 99% of the time they’ll say yes.
      - Dylan16807 4mo ago
        It doesn't matter what people call it. We're talking about maintaining a moat with extremely demanding use cases, and the extremely demanding range shrinks every few months.
        lazide 4mo ago
        I’m saying most of the moat was never there to begin with, and the rest is mostly immaterial to the bulk of the use cases.
squidbeak 4mo ago
> Otherwise it would be like Intel and Microsoft had decided in the year 2000 that computers are "good enough" now and we would have explored what's possible with that hardware ever since. I think you've misunderstood what good enough means in the context - which is a model capable of completing the tasks assigned to it without having the breadth of full generalization. Your analogy breaks down because of this - we did get 'good enough' spec profiles for different hardware. That thing you're wearing on your wrist won't have the same specifications as the box you use to play games.
- IdiotSavage 4mo ago
  I think you've misunderstood the analogy. Just ignore it, analogies mostly break down anyways. > a model capable of completing the tasks assigned to it The thing is, the "task assigned to it" is changing with improved capabilities. If everyone around you in 2036 is using general AI to do amazing stuff, you will probably have little interest in vibe coding slop like it's 2026.
  - brookst 4mo ago
    Analogies are like metaphors, they’re illustrative rather than literal.
  - coldtea 4mo ago
    >The thing is, the "task assigned to it" is changing with improved capabilities. Only if you give in to fads and FOMO. The core tasks people need change at a much smaller pace.
coldtea 4mo ago
>Otherwise it would be like Intel and Microsoft had decided in the year 2000 that computers are "good enough" now and we would have explored what's possible with that hardware ever since. That would be the dream... no fucking Electron! No lockdown modules.

benterix 4mo ago

> The AI companies will want to control what's possible and find new things to do that "need" their services. That's correct. The problem is they have smart people, tons of money, and several years to figure that out, and the best thing they can come up is a coding agent.

lazide 4mo ago
That isn’t the best thing they’ve come up with. It’s a marquee product that is fit for public consumption, however. The ‘best’ things are; - fuzzy pattern matching algorithms for traffic analysis, human and other image target recognition. - targeting algorithms that identify ‘suspicious’ individuals in large volumes of metadata. - fraud analysis - antagonistic image and video generation, both for fooling other fraud analysis, but also for propaganda, screwing with other actors, etc. - directed high speed content generation (text, pictures, video) to spam the ‘algorithm’ and allow near realtime identification of additional buttons to push for given target audiences. - massive marketing/ad manipulation. Those budget line items (and the suppliers) really want to stay off the radar however, as it makes their life harder.
- benterix 4mo ago
  But you're mentioning several things that predate the current LLM craze and belong to the ML domain. These mostly benefit from GPUs but often have much lower hardware requirements. I'm talking specifically about the moat of LLM providers.
  - lazide 4mo ago
    Sure, but all fall under the same marketing umbrella. Ad/marketing manipulation are exceptionally well done with LLMs in particular. If you asked someone if drone auto targeting/image recognition or data analysis was ‘AI’, 99% of the time they’ll say yes.
    - Dylan16807 4mo ago
      It doesn't matter what people call it. We're talking about maintaining a moat with extremely demanding use cases, and the extremely demanding range shrinks every few months.
      - lazide 4mo ago
        I’m saying most of the moat was never there to begin with, and the rest is mostly immaterial to the bulk of the use cases.

squidbeak 4mo ago

> Otherwise it would be like Intel and Microsoft had decided in the year 2000 that computers are "good enough" now and we would have explored what's possible with that hardware ever since. I think you've misunderstood what good enough means in the context - which is a model capable of completing the tasks assigned to it without having the breadth of full generalization. Your analogy breaks down because of this - we did get 'good enough' spec profiles for different hardware. That thing you're wearing on your wrist won't have the same specifications as the box you use to play games.

IdiotSavage 4mo ago
I think you've misunderstood the analogy. Just ignore it, analogies mostly break down anyways. > a model capable of completing the tasks assigned to it The thing is, the "task assigned to it" is changing with improved capabilities. If everyone around you in 2036 is using general AI to do amazing stuff, you will probably have little interest in vibe coding slop like it's 2026.
- brookst 4mo ago
  Analogies are like metaphors, they’re illustrative rather than literal.
- coldtea 4mo ago
  >The thing is, the "task assigned to it" is changing with improved capabilities. Only if you give in to fads and FOMO. The core tasks people need change at a much smaller pace.

coldtea 4mo ago