8 ms·

It’s not very soon, it’s already the case that if one wants to enable the latest models in the OpenAI api you have to submit your details to their “identity pro

by totetsu 7mo ago

It’s not very soon, it’s already the case that if one wants to enable the latest models in the OpenAI api you have to submit your details to their “identity provider”.

abracadaniel 7mo ago
Which is why it’s important to be able to run models locally. Which also might explain the strategy behind buying all of the memory that is or will exist for at least a year out. Maybe we’ll eventually see AI safety be used to prevent people from running local models.
- sandworm101 7mo ago
  +1 for local models. It also teaches users about how much energy they are using. One's perspective on 24/7 chatbots and agentic operating systems changes when you feel the heat coming from a rack of gpus. (Spring is nearly here and my excuse about my rig also heating my house is about to end. Soon I will be paying extra to run my a/c as my rig pumps out a steady 1000w under load.)
  - 1e1a 7mo ago
    You could use it to heat a tropical greenhouse.
- PeterStuer 7mo ago
  You mean having to sign into your Microsoft account to get your bootloader co-signed before your legally mandated TPU 3.1 allows you to install a govenment blessed and sufficiently telemetrized signed OS to "your" computer if you are on the whitelist of not-yet-misinformation-spreaders?
  - fennecbutt 7mo ago
    Well I suppose in that case it depends on how freedom loving TSMC (Taiwan) or ASML (Netherlands) want to be. No chips for you, random government. No chips for you either, or you. And you.
gruez 7mo ago
Given the recent mexican telecom hacks were allegedly done with significant help from openai/anthropic's chatbots, it seems at least somewhat prudent to require some sort of identity verification for API access? I'm struggling to see how this isn't the tech community's version of "no background checks for gun purchases" or "no KYC for bank accounts".
- paradox460 7mo ago
  They were also likely done with keyboards and mice. Should we require id at point of purchase for those?
  - gruez 7mo ago
    Alright, so does that mean we don't need KYC for gun purchases or bank accounts either? Of course you're probably going to say something about how guns and bank accounts are crucial components to crime, in which case the same holds for AI in the mexican telecoms hack.
    - BobbyJo 7mo ago
      What happens when everyone needs to use AI for their job? Genuine question that I think gets at the heart of the debate. Once a common technology that everyone has access to becomes powerful enough to alter the lives of others on command, do we as a society just need to do away with the concept of anonymity? We are all just too powerful in isolation, and too much of a threat to the collective, that we cannot reasonably expect not to have some governing body watching at all times? Today, you can buy parts/print a completely untraceable firearm, so do we license sales of steel tubing and 3D printers?
      - gruez 7mo ago
        >What happens when everyone needs to use AI for their job? Genuine question that I think gets at the heart of the debate. Considering most places does direct deposit and that requires a bank account (so KYC), I don't see what's particularly new here. Many places also do background and/or work eligibility checks, which again is a form of KYC. >Today, you can buy parts/print a completely untraceable firearm, so do we license sales of steel tubing and 3D printers? Fortunately 3d printed guns are bad enough that it's not really an issue, although the bigger threat is probably CNC machines. However that's probably will get a pass, because they're eye-wateringly expensive compared to black market guns that nobody would bother.

by totetsu 7mo ago

It’s not very soon, it’s already the case that if one wants to enable the latest models in the OpenAI api you have to submit your details to their “identity provider”.

abracadaniel 7mo ago

Which is why it’s important to be able to run models locally. Which also might explain the strategy behind buying all of the memory that is or will exist for at least a year out. Maybe we’ll eventually see AI safety be used to prevent people from running local models.

sandworm101 7mo ago
+1 for local models. It also teaches users about how much energy they are using. One's perspective on 24/7 chatbots and agentic operating systems changes when you feel the heat coming from a rack of gpus. (Spring is nearly here and my excuse about my rig also heating my house is about to end. Soon I will be paying extra to run my a/c as my rig pumps out a steady 1000w under load.)
- 1e1a 7mo ago
  You could use it to heat a tropical greenhouse.
PeterStuer 7mo ago
You mean having to sign into your Microsoft account to get your bootloader co-signed before your legally mandated TPU 3.1 allows you to install a govenment blessed and sufficiently telemetrized signed OS to "your" computer if you are on the whitelist of not-yet-misinformation-spreaders?
- fennecbutt 7mo ago
  Well I suppose in that case it depends on how freedom loving TSMC (Taiwan) or ASML (Netherlands) want to be. No chips for you, random government. No chips for you either, or you. And you.

gruez 7mo ago

Given the recent mexican telecom hacks were allegedly done with significant help from openai/anthropic's chatbots, it seems at least somewhat prudent to require some sort of identity verification for API access? I'm struggling to see how this isn't the tech community's version of "no background checks for gun purchases" or "no KYC for bank accounts".

paradox460 7mo ago
They were also likely done with keyboards and mice. Should we require id at point of purchase for those?
- gruez 7mo ago
  Alright, so does that mean we don't need KYC for gun purchases or bank accounts either? Of course you're probably going to say something about how guns and bank accounts are crucial components to crime, in which case the same holds for AI in the mexican telecoms hack.
  - BobbyJo 7mo ago
    What happens when everyone needs to use AI for their job? Genuine question that I think gets at the heart of the debate. Once a common technology that everyone has access to becomes powerful enough to alter the lives of others on command, do we as a society just need to do away with the concept of anonymity? We are all just too powerful in isolation, and too much of a threat to the collective, that we cannot reasonably expect not to have some governing body watching at all times? Today, you can buy parts/print a completely untraceable firearm, so do we license sales of steel tubing and 3D printers?
    - gruez 7mo ago
      >What happens when everyone needs to use AI for their job? Genuine question that I think gets at the heart of the debate. Considering most places does direct deposit and that requires a bank account (so KYC), I don't see what's particularly new here. Many places also do background and/or work eligibility checks, which again is a form of KYC. >Today, you can buy parts/print a completely untraceable firearm, so do we license sales of steel tubing and 3D printers? Fortunately 3d printed guns are bad enough that it's not really an issue, although the bigger threat is probably CNC machines. However that's probably will get a pass, because they're eye-wateringly expensive compared to black market guns that nobody would bother.