Well that's stupid. I submit though, connecting stochastic process directly to shell you do give permission for everything that results. It's a stupid game. Gemini mixes up LEFT and RIGHT (!). You have to check it.
I would submit once you obtain a certain level of experience it becomes IDEAL to begin with implementation, in case a mathematical analysis may be either trivial or impossibly non-trivial... Of course if you're dealing in exchange rates and risk management, understand the math!
The output distribution is altered - it starts responding "yes" 20% of the time - and then, conditional on that is more or less steered by the "concept" vector?
You're asking it if it can feel the presence of an unusual thought. If it works, it's obviously not going to say the exact same thing it would have said without the question. That's not what is meant by 'alteration'.
It doesn't matter if it's 'altered' if the alteration doesn't point to the concept in question. It doesn't start spitting out content that will allow you to deduce the concept from the output alone. That's all that matters.
They ask a yes/no question and inject data into the state. It goes yes (20%). The prompt does not reveal the concept as of yet, of course. The injected activations, in addition to the prompt, steer the rest of the response. SOMETIMES it SOUNDED LIKE introspection. Other times it sounded like physical sensory experience, which is only more clearly errant since the thing has no senses.
I think this technique is going to be valuable for controlling the output distribution, but I don't find their "introspection" framing helpful to understanding.
Either life really is extremely rare (most likely), or intelligent life is, or it isn't actually trivial/correct to destroy alien planets. If the galaxy were actually like that we would have been toast a long time ago. In reality dark forest is a generate thesis since it implies we're alone, so no aliens anyway.
We really don't know much about the universe and it is too vast and unfathomed. Scientists computed the mass of all matter and all energy of this Universe, but their calculations told that all this stuff comprises merely 5% of the Universe, the remaining 95% of the Universe is said to made of anti-matter and anti-energy, about which not much is understood.
So there's a good chance that aliens may be made of anti-matter and using anti-energy. But even if they tried to communicated with rest of universe with such anti-energy-based technology, we humans simply may not be detecting it or interpreting it yet, and we may still be waiting for that elusive signal (energy-based) indicating advanced intelligent life.
Nope, not "anti-". The 5% are visible "bright" mass and energy. "Bright" meaning that we can see it through telescopes, by various wavelengths of light, particle emissions, gravity waves. The rest is "dark matter" and "dark energy", which just means that we see signs of it being there, because the bright matter around it behaves differently. But we don't directly see it in a telescope of any kind. Those "dark" things are stand-ins for our not understanding: Those could be real matter and energy that we just cannot see for some reason. Or those could be problems in our cosmological theories, like gravity working differently on large scales, the expansion of the universe being different, or physical constants changing over time. We just notice that things are off and that we should see more matter and more energy than we do.
Most theories that involve "dark matter" being ordinary matter like tons of neutron stars, huge clouds of dust, bazillions of asteroids or dark planets have been checked for and excluded. So if there were "dark matter" aliens, they really would be completely strange in that they aren't even made from the same kind of matter, but from maybe particles that we don't even know about. But if those hypothetical dark matter particles were capable of this kind of organisation, like clumping together into stars or planets, we would have probably seen those by now. So extremely strange, and improbable imho.
Btw. anti-matter is not "dark matter" in this sense, and dark matter being anti-matter was excluded very very early on by a simple observation: anti-matter and matter, when they come into contact, react in an annihilation reaction. E.g. an electron and anti-electron annihilate into two photons of a characteristic and exact 511keV energy. All other particles and their anti-particles also do this and exhibit their own characteristic energy. Any contact between a region of matter and region of anti-matter in space would radiate in these energy signatures, something which is very easy to detect. Dark matter is known to exist within galaxies, even within star systems, so this kind of contact zone would have to be there, and would be extremely visible to us.
Anti-energy doesn't exist in our current understanding of physics. Energy is always positive, and in quantum theories energy cannot even become zero, always slightly above zero.
The fundamental driver of the economy will be people, as always, because only people define "value".
Word generating machine will not "figure out how to cure cancer" but it could help, obviously. AI is extremely valuable tool but it does not work on my behalf, except in same sense a coin sorter does. It's a tool. I still see this thing confuse left and right (no exaggeration), which would be fine - tools aren't perfect - except for all bullshit from VCs. That is where the danger lies, not with the tool but idolatry.
I am concerned the system encourages suicide, delusional thinking etc. They need to work that out immediately. Must be held to at least the standard of a lawn mower or couch with regard to safety. Probably should make it safe before hooking it up to 10GW power plant. Does not help perception that author of this blog saying how awesome future will be is also building a hideaway bunker for himself.
This is all on you at OpenAI, Anthropic, Microsoft, Twitter etc. Whatever happens.
I didn't think reasonable really has a definition here. The rendezvous with these comets will take centuries or millennia even if we can get out there and kick them the right way (which is much harder than arranging a miss). Only then do we start terraforming, which takes thousands/millions of years in addition. Who/whatever we're preparing the planet for probably wouldn't be "people" anymore. :)
> Only then do we start terraforming, which takes thousands/millions of years in addition
The methods we could realistically launch into in our lifetimes would take thousands of years, not millions (but also not hundreds) [1]. Projects of these timescales have precedent in human history, usually with a healthy dose of religious zeal.
My impression was that pulverized rock/iron is not actually "soil", so after forming the atmosphere you need lengthy biochemical processes playing out on the surface. I admit though, I don't know too much about this.
I think that we should consider "the future". Yes, it's intangible but consider this; go back 2 centuries and ask someone if they could setup a business concern which produced millions of widgets a year.
They'd think you daffy.
Now beyond that, ask them to produce any manner of modern device with the precision and high consistency we have. Again, they'd think you mad, and think that such was impossible.
Yet here we are.
The next stage in our development via LLMs is not about AI helping humans. It's about robotics. Automated assembly. Robots (not Androids) able to interact with the environment and able to problem solve akin to say.. a mouse.
Soon, entire factories will be entirely automated. Many almost are. We don't need Von Neumann machines to see this future, but we will certainly have robots capable of building entire factories, collecting resources and processing them, and further building machines to spec. And those machines will be able to self-drive, self-operat autonomously.
Anyone playing typical resource games knows about bootstrapping, but once in the asteroid field we're basically resource infinite. Building engines to attach to asteroids, mining asteroids, building factories to create more robots and engines, all of it will be automated.
We toil at self-driving cars, yet this same tech enables self-driving robotics of all types.
So I honestly think that once we bootstrap in space, this sort of thing can happen fast, fast, fast. Decades to send hundreds of thousands of ice-rich resources to Mars.
The soil? Ah, genetic engineering. Really, this is an entirely new field, and frankly is beyond the danger yet benefit of nuclear science. We have the bomb, yet we have nuclear energy and medicine. Well genetics can obviously be far more deadly, and research all over the world, and startups, are already working on employing bacteria and organisms as self-replicating machines to do our bidding.
The dangers are in our face, but oh well! So if we presume survival, then once an atmosphere is produced we can seed the planet with organisms which can survive on rock and yet work with a mania to process it. It's OK if we immediately have moss like grass substitute everywhere. As long as it's working its magic, we get continued O2 production, and we can always create a rabbit pet or something that licks moss to survive. Or are tasty.
My point is, there are indeed many barriers. But we need to view them with where we will be in decades, not where we are now.
> Yes, it's intangible but consider this; go back 2 centuries and ask someone if they could setup a business concern which produced millions of widgets a year.
To go off on a tangent: two centuries ago was the height of the first industrial revolution (at least in Britain). The first time in history when this actually became realistic.
The Industrial Revolution was the first time we had sustained, broad based productivity growth year after year (even if only around 1%, which is quite low by modern standards).
Weirdly enough, we can see sustained productivity growth in artillery and guns long before the wider industry.
Another weird connection: sometimes people look at a toy 'steam engine' that the ancient Romans had access to (https://en.wikipedia.org/wiki/Aeolipile) and wonder if they could have had an industrial revolution. But, to make a proper steam engine you need a lot more than just the right idea. You need a lot of metallurgy and precise crafting.
Specifically one thing you need is precision crafted cylinders that gas can expand in to move a piston. Well, at the time of the Industrial Revolution, European nations had just spent several hundred years locked in existential competition over who can make precision crafted cylinders that gas can expand in to move a bullet.
I wonder though, if not it would have been possible to build stationary steam engines with Roman tech using oversized bronze castings for cylinders. Perhaps set in bedrock to give extra strength.
Weirdly though, electric generators in watermills would have been much more attainable - except nobody had any understanding of electricity.
Steam engines were stationary at first. They were used to eg drive pumps.
> Weirdly though, electric generators in watermills would have been much more attainable - except nobody had any understanding of electricity.
Yes, and proper dynamos were invented only quite a long time after batteries. (So called self-excited generators.)
And you have to compare the early bad electric generators they could have come up with against the gears and shafts they knew to transmit the motive force of the water over short distance eg to the mill stone.
You make a lot of good points. Including that the technology to do all that incredibly dangerous to develop - I still can't see a path to terraforming where in the end it's human people left to take pleasure in Mars.
I had thought due to the eons we'd simply have evolved, but even on shorter time frames there is the transhumanist possibility. When we can engineer rabbit that eats chlorine moss, I don't know what we're aiming for at all. "People" by then could have robust gut culture that just digests the regolith.
There's a difference between considering all this vs thinking it's realistic. It's speculation, as any forecast into the centuries ahead must be.
I recommend The Image by Daniel Boorstin. Smart phone is not the dawn at all but I guess more like high noon. The world is replaced by graphics (more generally , images) for a century at least. Books by "digests", heroism by celebrity and so forth. A little bit of media theory goes a long way to making the world "legible" again. Since we are absolutely steeped in media culture.
Fahrenheit 451 didn't get as embedded into popular culture as 1984, but does a good job depicting a society brain rotted on reality TV and war propaganda.
"the world has been replaced by graphics (more generally, images)"
Humankind started to record its history by images (google for instance about the city of Sefar, Algeria). Nowadays, even in tech we use graphics (diagrams and so on)
This is not my field but even the first letter of the latin alphabet is simply the image of the head of a cow rotated a bit to the left.
But working with complex numbers I hardly if ever write (a, b) for a+ib, while I use the "escape hatches" all the time. They solve equations that have no real solution, they give me paths from x=-1 to x=1 that don't cross the origin, etc. There's only so much to learn about C as a vector space, while the theory tying it to R (and even N) is very deep.
Thing is, there's no such thing as an escape hatch. Either you are working in the reals, or you are working in the complex plane. They don't "solve equations that have no real solution", that equation is either a real number equation or a complex number equation, not both. If you work in the complex plane, that is a different equation describing a different space! It just looks the same in standard notation.
If you don't realize this, then you can draw conclusions that don't make sense in the space you're working with. Take a simple equation like y = -x^2 - 5, representing a thrown ball's trajectory. It never crosses zero, there are no solutions. You can't "pop into the complex numbers and find a solution" because the thing it represents is confined to the reals.
So if you find yourself reaching for complex numbers, you have to understand that the thing you are working with is no longer one-dimensional, even if that second dimension collapses back to 0 at the end.
reply