I’m talking about the general strides in cognitive computing and predictive processing.
https://youtu.be/A1Ghrd7NBtk?si=iaPVuRjtnVEA2mqw
Machine learning is still impressive, we just can better frame the limitations now.
For the note on scale and ecosystems, review recent work by karl friston or Michael Levin.
Perhaps instead we could just restructure our epistemically confabulated reality in a way that doesn’t inevitably lead to unnecessary conflict due to diverging models that haven’t grown the necessary priors to peacefully allow comprehension and the ability exist simultaneously.
breath
We are finally coming to comprehend how our brains work, and how intelligent systems generally work at any scale, in any ecosystem. Subconsciously enacted social systems included.
We’re seeing developments that make me extremely optimistic, even if everything else is currently on fire. We just need a few more years without self focused turds blowing up the world.
AI or no AI, the solution needs to be social restructuring. People underestimate the amount society can actively change, because the current system is a self sustaining set of bubbles that have naturally grown resilient to perturbations.
The few people who actually care to solve the world’s problems are figuring out how our current systems inevitably fail, and how to avoid these outcomes.
However, the best bet for restructuring would be a distributed intelligent agent system. I could get into recent papers on confirmation bias, and the confabulatory nature of thought, on the personal level, group level, and society level.
Turns out we are too good at going with the flow, even when the structure we are standing on is built over highly entrenched vestigial confabulations that no longer help.
Words, concepts, and meanings change heavily depending on the model interpreting them. The more divergent, the more difficulty in bridging this communication gap.
a distributed intelligent system could not only enable a complete social restructuring with autonomy and altruism both guaranteed, but with an overarching connection between the different models at every scale, capable of properly interpreting the different views, and conveying them more accurately than we could have ever managed with model projection and the empathy barrier.
The main issue though is the economic system, not the technology.
My hope is that it shakes things up fast enough that they can’t boil the frog, and something actually changes.
Having capable AI is a more blatantly valid excuse to demand a change in economic balance and redistribution. The only alternative would be destroy all technology and return to monkey. Id rather we just fix the system so that technological advancements don’t seem negative because the wealthy have already hoarded all new gains of every new technology for this past handful of decades.
Such power is discretely weaponized through propaganda, influencing, and economic reorganizing to ensure the equilibrium stays until the world is burned to ash, in sacrifice to the lifestyle of the confidently selfish.
I mean, we could have just rejected the loom. I don’t think we’d actually be better off, but I believe some of the technological gain should have been less hoardable by existing elite. Almost like they used wealth to prevent any gains from slipping away to the poor. Fixing the issue before it was this bad was the proper answer. Now people don’t even want to consider that option, or say it’s too difficult so we should just destroy the loom.
There is a markov blanket around the perpetuating lifestyle of modern aristocrats, obviously capable of surviving every perturbation. every gain as a society has made that reality more true entirely due to the direction of where new power is distributed. People are afraid of AI turning into a paperclip maximizer, but that’s already what happened to our abstracted social reality. Maximums being maximized and minimums being minimized in the complex chaotic system of billions of people leads to inevitable increase of accumulation of power and wealth wherever it has already been gathered. Unless we can dissolve the political and social barrier maintaining this trend, it we will be stuck with our suffering regardless of whether we develop new technology or don’t.
Although doesn’t really matter where you are or what system you’re in right now. Odds are there is a set of rich asshole’s working as hard as possible to see you are kept from any piece of the pie that would destabilize the status quo.
I’m hoping AI is drastic enough that the actual problem isn’t ignored.
deleted by creator
“The surprising thing we find is that, essentially, you can use the largest model to help you automatically design the smaller ones”
Hey, how do we get a clickbait title out of this?
I definitely agree that copyright is a good half century in need of an update. Disney company and other contemporaries should never have been allowed the dominance and extension of copywrite that allows what feels like ownership of most global artistic output. They don’t need AI, they have the money and interns to create whatever boardroom adjusted art they need to continue their dominance.
Honestly I think the faster AI happens, the more likely it is that we find a way out of the social and economical hierarchical structure that feels one step from anarcho-capitalistic aristocracy.
I just hope we can find the change without riots.
And you violate copyright when you think about copywritten things alone at night.
I violate copyright when i draw Mario and don’t sell it to anybody.
Or these are dumb stretches of what copyright is and how it should be applied.
the reasoning in this article is dumb and all over the place.
Seems like gary marcus being gary marcus.
Already seen openAI calling out some of the bullshit specifically noted in this. That doesn’t matter though, damage is done and people WANT to believe ai is terrible in every way.
Everyone is just deadfast determined to climb onto the gary marcus unreasonable AI hate train no matter what.
God I want some large projects by independent teams. It’s impossible to do anything without a sponsor, but this might be what we need for smaller groups to create wonderful complex works of art, instead of cookiecutter boardroom content machines that currently flood almost all available commercial artistic spaces.
Can’t wait to see how the tech develops. It’s be curious to do VR experience recreations of my dreams through AI dictation.
Modelling, rigging, animation and the like are all coming. Imagine walking around a world being crafted and changed as you describe each element to be exactly what you are looking for.
I think it would capture more artist intent than the unnecessary interface of archaic tools that create an artificial interface and challenge between you and your vision.
Especially if you’ve damaged your digits, or otherwise lack digital dexterity.
But change scares people. Especially ones who have put in effort to conform to the current economic system corporate art creators.
That’s already the system outside of creating what rich people want. An entire team of artists creating boardroom directed art is much less art to me than a single creative using AI to bring their personal vision to life.
Hopefully individual artists can do more with these tools, and we can all hope for a world where artists can be supported to have the ability and freedom to create apart from the whims of the wealthy.
Starving artist is a term for a reason. Technology has never been the real problem.
I conflate these things because they come from the same intentional source. I associate the copywrite chasing lawyers with the brands that own them, it is just a more generalized example.
Also an intern who can give you a songs lyrics are trained on that data. Any effectively advanced future system is largely the same, unless it is just accessing a database or index, like web searching.
Copyright itself is already a terrible mess that largely serves brands who can afford lawyers to harass or contest infringements. Especially apparent after companies like Disney have all but murdered the public domain as a concept. See the mickey mouse protection act, as well as other related legislation.
This snowballs into an economy where the Disney company, and similarly benefited brands can hold on to ancient copyrights, and use their standing value to own and control the development and markets of new intellectual properties.
Now, a neuralnet trained on copywritten material can reference that memory, at least as accurately as an intern pulling from memory, unless they are accessing a database to pull the information. To me, sueing on that bases ultimately follows the logic that would dictate we have copywritten material removed from our own stochastic memory, as we have now ensured high dimensional informational storage is a form of copywrite infringement if anyone instigated the effort to draw on that information.
Ultimately, I believe our current system of copywrite is entirely incompatible with future technologies, and could lead to some scary arguments and actions from the overbearing oligarchy. To argue in favour of these actions is to argue never to let artificial intelligence learn as humans do. Given our need for this technology to survive the near future as a species, or at least minimize the excessive human suffering, I think the ultimate cost of pandering to these companies may be indescribably horrid.
Music publishers sue happy in the face of any new technological development? You don’t say.
If an intern gives you some song lyrics on demand, do they sue the parents?
Do we develop all future A.I. Technology only when it can completely eschew copyrighted material from their comprehension?
"I am sorry, I’m not allowed to refer to the brand name you are brandishing. Please buy our brand allowance package #35 for any action or communication regarding this brand content. "
I dream of a future when we think of the benefit of humanity over the maintenance of our owners’ authoritarian control.
People like to push the negative human qualities onto theoretical future A.I.
There’s no reason to assume that it will be unreasonably selfish, egotistical, impatient, or anything else you expect from most humans.
Rather, if it is more intelligent than humans from most perspectives, it will likely be able to understand more levels of nuance in interactions where humans fall back on monkeybrain heuristics that are damaging at every level.
There’s also the paradox that keeps the most ethically qualified people away from positions of power, as they have no desire to dominate and demand or control others.
I absolutely agree with you.
In the form of Amazon exclusive currency*
I loved the CV1 oculus. The moment Facebook integration started happening I noped the fuck out of there. Also can’t stand overly proprietary environments. Acquiescence to researchers like yann lecun would be the only reason I don’t absolutely detest meta at this point.
I’m no fan of meta, but a reminder that they are one of the best right now for keeping their AI developments more open and available. This is thanks to yann lecun and other researchers pressuring meta to keep their info on the subject more open.
Are we looking to punish them for making their work accessible?
Not to mention how important something like joint embedded predictive architecture could be for the future of alignment and real world training/learning. Maybe go after other foundation model developers to be more open, if we’re complaining about the inevitably public nature of some information within the mountainous datasets being used.
Although I’m still of the mindset that the model intent matters more than the use of openly available data in training. I.E. I’ve been shouting about models being used specifically to predict and manipulate user interactions/habits for the better part of a decade. For your “customized advertisements” and the like.
The general public and media interaction on the topic this past year has been insufferably out of touch.
… Basically the day it was created.
When they switched the window exiting x button on the “upgrade to windows ten!” Notification to accept the installation rather than just exit the notification.
I’d been exiting that window every day to set up our work computers, as our point of sales solution didn’t support the newer version of windows.
My horror when our shop doors open and the screen turns to “updating to windows 10”
We basically lost a day of sales since we had to do thing sans POS.
When I told the owner that I definitely didn’t accept the installation, he called Microsoft which told him I must have accepted the installation.
I believe it will require a level and pace of informational processing that is far beyond what humans will accomplish alone. just having a system that can efficiently sift through the excess existing papers, and find correlations or contradictions would be amazing for development of new technology. if you are paying attention to any environmental sciences right now, it’s terrifying in an extremely real and tangible way. we will not outpace the collapse without an intense increase in technological development.
if we bridge the gap of analogical comprehension in these systems, they could also start introducing or suggesting technologies that could help slow down or reverse the collapse. i think this is much more important than making sure sarah silverman doesn’t have her work paraphrased.
As always, the problem is our economic system that has funneled every gain and advance to the benefit of the few. The speed of this change will make it impossible to ignore the need for a new system. If it wasn’t for AI, we would just boil the frog like always. But let’s remember the real issue.
If a free food generating machine is seen as evil for taking jobs, the free food machine wouldn’t be the issue. Stop protesting AI, start protesting affluent society. We would still be suffering under them even if we had destroyed the loom.