Thing is, we don’t understand how LLMs work. Not really. And everyone at the forefront of this space has said so. We know how to grow them but that is not anything like the same thing as understanding why or how they actually function. Our level of understanding of LLM intelligence is only slightly ahead of our understanding of the neuroscience of human intelligence, and no one would say we understand that at all.

So I think describing LLMs as magic at this point is absolutely appropriate.

  > very different from what humans would call conscious

I think the real issue is that when most people refer to consciousness, they have their own subjective experience in mind which strongly resists any tidy definition. I think it’s extraordinarily unlikely LLMs have anything like this, but they are far more able to effectively respond to their surroundings than most animals and in some areas better than humans.

So if you’re waiting for proof that an LLM has an inner life basically equivalent to your own, you’ll be waiting a long time. After all, other humans can’t even prove the fact of their own consciousness to you! They could just be replaying their training data at you in a way that is merely a convincing but false simulation of the true consciousness which you experience inside your head.

  >  Is it even possible to reliably know that a connection is from a VPN?

Comments URL: https://news.ycombinator.com/item?id=49928904

Points: 12

# Comments: 2

  > if it takes an unsigned commit and signs only its SHA-1 hash and then creates a new commit with GPG headers

  tree eebfed94e75e7760540d1485c740902590a00332
  parent 04b871796dc0420f8e7561a895b52484b701d51a
  author Alice  1465981137 +0000
  committer Alice  1465981137 +0000

  Headline

  Message

> most of their time waiting for the models response and tool calling rather than running their code.

You’d think that! Yet claude-code spends a very surprising amount of CPU just doing text layout work and other mysterious things, likely due to their decision to use React to build a TUI for some reason.

More seriously: the people I’m objecting to are those that constantly chime in to unrelated conversations when someone says eg “the LLM thinks that X” with “please don’t anthropomorphize, LLMs do not think”. They are the ones policing language, it’s self-evidently silly to call my objection to their behavior yet another attempt to police language. They can use whatever language they want in their own conversations, I want them to stop jumping in to conversations they are otherwise involved in with no goal other than to scold. Surely you can see how there’s a difference?

More substantively: we don’t have a real definition for most of these anthropomorphic terms like “think” and “feel” and “want” and they are the closest match to the observed behavior of LLMs in most cases. It’s pointless to object to the way people naturally use language.

Uh, a hypothetical fully open source model would have the exact same problem, because LLMs rely on emergent phenomena and no one understands why they work.

But I do have a problem generally with attempts to impose language changes on others (in the case of this post you only see that tendency in the title).

This sort of linguistic prescriptivism seems in vogue for political reasons lately, most saliently in the hordes of people who insist that you must not use natural anthropomorphic language when talking about LLMs, but that’s far from the only political attempt to control language we’ve seen recently. Let people talk how they want! Langauage is an evolving thing that shouldn’t be controlled by scolds!

2. What’s your point? People are demanding justice for behavior they don’t like but isn’t illegal? The way to fix that is to gather a majority and change the law.

My company uses Claude models exclusively via Azure and AWS bedrock, which have their own licensed copies of the weights.

If all these people are so convinced Anthropic is nerfing models, have they tried other inference providers? Do they think the nerfing is coordinated across independent providers? Why wouldn’t any of these nerf-benches use these comparison points or even talk about them?

I think the most likely conclusion by far is that this is a psychological phenomenon.

I thought we were supposed to be hackers. Information wants to be free and all that. What a bunch of hypocrites, willing to betray their ideals as soon as it’s convenient.