Nearholmer wrote: 26 Aug 2025, 10:58am
Experience of absorbing extra-somatic information isn't a full or true experience of the world, though, is it?
Of course not, but it leapfrogs a person ahead very effectively in a huge number of applications, they don’t have to work out from scratch for themself the entire body of scientific knowledge and philosophical insight. In the former case, and maybe the latter too, they can stand on the shoulders of giants.
As to how people, and the machines that people create, become able to synthesise oodles of knowledge and then make use, good, bad, or indifferent, of it to inform actions, Im not entirely sure, but people certainly do, tweaking their responses in accordance with what works and what doesn’t (learning from experience), incorporating new knowledge as they go etc.
It’s all very easy to pour doom and scorn on the human capacity to reach workable decisions on the basis of oodles of real-time inputs and stored knowledge, and to say that it takes us to bad and dark places (which it unquestionably does sometimes), but it’s proven itself very amply indeed by allowing us to proliferate. It works, and I’d wager that we can create machines in which it works …… the key question is “if, as it seems, our prime directive is to proliferate”, then what should we make theirs? Simply telling them to go forth and multiply might not be a very wise idea.
Non-somatic information and its use can, as you say, "leapfrog a person ahead very effectively" in various domains of specialised human doings. The harder sciences are perhaps the best example of this, not least because a lot of the knowledge involved is mostly used to derive further such knowledge, with only a modicum translating into more practical doings. In reality, though, its often the experience of performing practical endeavours that generate theoretical knowledge or improvements to it in the first place.
Technology, historically, has often come before the various scientific understandings that can clarify and improve that technology. In fact, the scientific method recognises this, generally claiming that the process begins with evidence and fact collecting, which then tends to suggest theories of why things work as they do. (In reality, of course, cultural pre-conditions often suggest a theory first, with evidence and facts collected or formed to bolster that theory, sometimes ignoring evidence and facts that don't).
But there are many academic domains, with large bodies of "scientific knowledge and philosophical insight" that are a very poor guide to reality; and not very effective at all at predicting the future or determining the minutiae of how something actually works or doesn't work. Economics is one good example. Psychiatry, sociology and political philosophy are other such examples. What a student of these leapfrogs into is often a domain of SNAFU and ineffective techniques, some doing serious harm.
*************
The dominance of non-somatic learning is rather a new thing in human history. It's only come to the degree of dominance and effect on history in the last few thousand years, with the greatest effects in the last 250 years. Not that long ago the vast majority of human learning was via practice, not theorising. Humans did things rather than writing or reading about them.
It is the case that modern education and media have vastly accelerated many of the practical intents often claimed as great human successes. But the effects have not yet fully come to fruition. Whilst the human population has increased dramatically via modern medicine, agriculture and industry, many of that increased population live miserable and degraded lives as a result, even in First World countries like ours. Moreover, the full effects of all those clever non-somatic learnings and their translation into real world doings are becoming more apparent; and they do bode a great ill for not just humans but a vast swathe of the biosphere, in which many species have already gone extinct in a very short time due to our clever doings.
***************
So, even we clever humans, with our mix of theoretical and practical knowledge & skills, have failed to act in a truly intelligent way. AI that has no recourse but to the kind of knowledge and understanding that's non-somatic may have even more of an inclination to wreak unintelligent acts into the world. Some will be seeded by human stupidities and some may be native stupidities generated by the AIs themselves.
*************
AIs made able to directly act in the physical world and measure/evaluate their experiences there seem likely to be both the best hope for a truly intelligent AI .... but also a serious existential risk to the biosphere, especially humans. Robotic entities driven by AI seem likely to be at the bleeding edge of a more versatile AI development.
Self driving cars, the more optimistic humans dream of. Terminator-like hunter-killer drones are the entities most feared by the pessimists, of which I am a-one. These things are being rapidly developed in the Russia-Ukraine war, copied by malign states such as the US and Israel and are likely to eventuate in a fully-developed form as a automated factory producing and emitting millions of things, each with a neural-network learning capacity enabling them to get very good at hunting and killing humans with various attributes seeded into the drone AI by the warring humans.
But we may first all be dead of weather and the consequent vast changes climate change is bringing to food production, trading practices and several other human systems highly vulnerable to any significant perturbation. All wrought courtesy of non-somatic learning and its leapfrogging ...... to our doom. Maybe the biosphere would have been better served if we'd remained in the stone age?
“Practical men who believe themselves to be quite exempt from any intellectual influence are usually the slaves of some defunct economist”.
John Maynard Keynes
