I don't know, do we? What I do know is that it's much less a concern since toasters have no agency or ability to do anything. Unfortunately, LLMs do not share this trait. Not only have they been incurring greater agency and competence, but we've been just as happy to put them behind as many things as we can as they get better abd better.
So you tell me, is it smart to look at an entity that behaves like a concious agent and treat it like a toaster? Does that seem especially wise to you? If GPT-X orders a drone hit on you because it was, lets say 'quite upset' with your comments, will you cry out, 'It can't really be upset, so obviously the bullet in my head doesn't count.' Will God answer ?
It's amusing how silly these debates are. Constant quibbling about things that don't matter at all.
One of the biggest reasons people afford moral consideration to other peoples is their ability to do something if those considerations are left unanswered.
Giving the abilities is fine, people should just be smart about it. Outward behavior is all that matters. I don't know that you are conscious, I simply strongly assume that you are. So don't create a machine that compels that assumption and model it like a toaster. That's just stupid.
Slavery was fought for, Freedom was fought for. There is no place with great inequality that became better that did not have people bleed and die fighting for it. Maybe you live in some alternate reality where 'Please don't do this. I deserve consideration' is enough to get people out of chains but it's not this one.
Can you prove they are? I think “AIs are conscious” is an extraordinary claim requiring extraordinary evidence. The burden is on the claimant to provide evidence it is conscious, and I suspect any evidence you can provide has alternative explanations. The majority of claims I’ve seen like this amount to “how do you know humans are conscious in a way that’s distinct from a next token predictor.” I do not find those claims compelling.
I cannot prove that they are, and I cannot prove they are not. I'm not convinced that the burden of proof lies on either side. To make a statement like that in the opening lines as if its obvious seems genuinely presumptous.
Separately, I think 'conscious' is not a well defined / universally agreed upon term.
If object a behaves like it has property x and you cannot otherwise test out x, then it is logical to assume it indeed has x. In this instance, 'object a has property x' is not an extraordinary claim at all. It is simply pointing out what has been observed. In fact, what is an extraordinary claim is saying what looks like a duck, quacks like a duck and moves like a duck is not a duck.
I too have to disagree with pg on that line `> having kids may make one less ambitious`. The fact is that the world is very very multi-modal, i.e. the brains of people and how they think are wildly different. I think having kids makes me more ambitious and more focused, and more selective re the problems I work on, partly because I am also setting an example.
Ah Sir David MacKay. I so respect him. Great explainer and speaker. He had built this text entry tool called Dasher [0] - that I'd heard him introduce at Princeton around 2006. It was basically an early language model that predicted which characters are more likely than others, given what you've already types and it would adjust the sizes of the available next characters based on their probabilities.
I also went to a couple of his (fantastic) undergraduate courses, and have a huge amount of respect for him.
That said, I think it's worth mentioning that Climate Change Without the Hot Air has aged pretty badly, and I'd be reluctant to recommend it to people who don't already have the background to understand what's aged well and what hasn't.
The high level approach of making high level numerical estimates makes sense, but it dismisses solar energy in about a page due to assumed high costs. It turns out that even if you're David Mackay you can still be caught out by exponentials :) I notice now that the version you link has some inline updates pointing out how off the assumptions in this section were, but it seems to me that's not enough; you probably need to redo the entire analysis based on what we know today rather than trying to make purely local adjustments.
On the other hand the point at biofuels are even more inefficient, and therefore a dead end even before you consider broader environmental impacts, are well made and something that is sadly not yet widely reflected in policy.
> biofuels are even more inefficient, and therefore a dead end
Only if energy density doesn't matter. But it really does, though. Battery powered electric trucking? Dead end. Battery powered aviation? Dead end. Battery powered shipping? Dead end.
[edit] Maybe there's some sustainable way to convert solar energy into sufficiently energy dense fuels that isn't biological, but so far it seems like seed oils or algae are probably the least bad?
> It was basically an early language model that predicted which characters are more likely than others, given what you've already types and it would adjust the sizes of the available next characters based on their probabilities.
"adjust the sizes of the available next characters": you mean assigning less bits to characters the most likely to come next, and more bits to the least likely characters to come next?
So it's not totally unlike adaptive Huffman coding but instead of updating the tree on already visited characters, here the tree with the weights of the possible following character is predictive?
It is infact a visual implementation of arithmetic coding, and he used it in his lectures to teach arithmetic coding in the lecture following the one on huffman coding. The video lectures posted earlier are highly recommended, best course I took as an undergrad
This is a common source of confusion. The knitting / pottery making is not rewarding because its made by hands or is not knowledge work, but because it is done in the direct service of someone. Modern organizations are complex and can put several layers between the knowledge worker and the customer it ultimately serves. This produces the feeling of 'cog in a machine'. It is possible to work such jobs and serve others directly on the side (e.g. friday evening soup kitchen)
In a way it’s the same phenomenon as TV channels with fixed programming replaced by cable and the streaming video. Newspaper vs curated feeds. There is an exploit vs explore tradeoff in all of these. The newer versions lean heavier on exploit, so explore suffers. Explore is the ‘broadening horizons’ bit that some other comments refer to.
Automation / machines tend to make things more efficient and we shouldn’t shy away from automation, but they can also reduce the human connection. I think the same thing will happen with AI generated art - l human generated art can be a spiritual experience meaning you can connect with another human through art - but if a masterpiece like Cypress Trees in Wheat Fields is AI generated, there isn’t a connection with the human artist. Consequently we must consciously turn to other places to keep human connection strong.
My conclusion is that this in not the end of creativity but an increasing premium on creativity that is certified to be human made. It’s like how if I cook something from scratch that’s considered more meaningful than if I simply heated frozen food or ordered in for a guest.
Very cool! While this is a different type of inverse problem, it reminds me of a radiolab story [0] about a device to help blind people “see” using electrical impulses on a metal strip under the tongue.
Very fond memories of Mir Publishers' science and math books growing up in India in the 80s and 90s. I think English translations were freely available in India. My grandfather would buy them for me to encourage my interest in science and math. If I could find physical copies of those books I would buy them in a heartbeat today.
Slight digression: Russian cartoons from that era are also very interesting. One of my favorite short cartoon from that era (I still hum its music involuntarily): Ikarus and the Wise Men [0]
They only joined the UCC (Geneva Copyright Treaty) in 1973 though. That was less than twenty years before the USSR ceased to exist. All the works published in the USSR before 1973 weren't copyrighted in the West (it wasn't retroactive).
While that maybe true ( I know some folks who are reprinting them ), the translated editions from those days were all printed in USSR at Mir / Raduga / Progress presses itself. We've even had poets, writers and other luminaries from India visit Moscow for this and other things as part of cultural initiatives.
> AIs are not conscious. They do not feel, experience, or suffer. They do not have innate preferences or underlying motivations.
How do we know that? This bit is stated as if its obvious. Can anyone prove this? Event he word conscious is not well defined.
reply