I imagine our world would be mostly incomprehensible to someone from the 1400s (the lack of centricity of religion, assuming some infernal force is keeping airplanes aloft, etc., to say nothing of the internet). If superintelligent AI really does take over the world, I image the most uncomfortable part of it all will be explaining to future generations how we were just too lazy to stop it.
Assuming climate change doesn't get us first.
You mean (we) got algorithmic feeds which feed us antisocial content for the sake of profit because such content drives the most engagement thus generating the most ad revenue.
Yes they do, but in order to do that, LLMs soak up the statistical regularities of just about every sentence ever written across a wide swath of languages, and from that infer underlying concepts common to all languages, which in turn, if you subscribe at least partially to the Sapir-Wharf hypothesis, means LLMs do encode concepts of human cognition.
Predicting the next token is simply a task that requires an LLM to find and learn these structural elements of our language and hence thought, and thus serves as a good error function to train the underlying network. But it’s a red herring when discussing what LLMs actually do.
I listened to a radio segment last week where the hosts were lamenting that Europe was able to pass AI regulation but the US Congress was far from doing so. The fear and hype is fueling reaction to a problem that IMO does not exist. There is no AI. What we have is a wonder of what can be achieved through LLMs but it's still a tool rather than a being. Unfortunately there's a lot of money to be made pitching it as such.
What are we humans even doing when zooming out? We're processing the current inputs to determine what best to do in the present, nearest future or even far future. Sometimes, in a more relaxed space (say a "brainstorming" meeting), we relax our prediction capabilities to the point our ideas come from a hallucination realm if no boundaries are imposed. LLMs mimic these things in the spoken language space quite well.
AND
> ... do encode structural elements of our language and hence thought
Quite true. I think the trivial "proof" that what you are saying is correct is that a significantly smaller model can generate sentence after sentence of fully grammatical but nonsense sentences. Therefore the additional information encoded into the network must be knowledge and not syntax (word order).
Similarly, when there is too much quantization applied, the result does start to resemble a grammatical sentence generator and is less mistakable for intelligence.
I make the argument about LLMs being a time series predictor because they happen to be a predictor that does something that is a bit magical from the perspective of humans.
In the same way that pesticides convincingly mimic the chemical signals used by the creatures to make decisions, LLMs convincingly produce output that feels to humans like intelligence and reasoning.
Future LLMs will be able to convincingly create the impression of love, loyalty, and many other emotions.
Humans too know how to feign reasoning and emotion and to detect bad reasoning, false loyalty, etc.
Last night I baked a batch of gingerbread cookies with a recipe suggested by GPT-4. The other day I asked GPT-4 to write a dozen more unit tests for a code library I am working on.
> just about every sentence ever written across a wide swath of languages
I view LLMs as a new way that humans can access/harness the information of or civilization. It is a tremendously exciting time to be alive to witness and interact with human knowledge in this way.
Why would anyone subscribe to the Sapir-Wharf hypothesis, in whole or in part?
As a society we've been dedicating a lot of resources to time series prediction for many years.
What makes LLMs culturally significant is that they generate sequences that map to words that seem to humans like intelligent responses.
Arguably, it has always been obvious that a sufficiently capable time series predictor would effectively be a super-weapon.
Many technological advances that are currently in the realm of sci-fi could be classified similarly.
However so could many technologies that are now widespread and largely harmless to the status quo.
People worried that the internet would create massive social upheaval. But soon got algorithmic feeds which effectively filter out antisocial content. The masses got mobile phones with cameras, but after a few scandals about police brutality the only place we find significant content about police misconduct is CCP-afiliated TikTok.
I think people get squeamish about AI because there are not clear authority structures other than what one can buy with a lot of A100s. So when people express concern about negative consequences, they are in effect asking whether we need yet another way that people can convert money + public resources into power while not contributing anything to society in return.