seydor

Reinforcement Learning with Divine Feedback

Ultra Pro Max Intelligence 2.7

Much of europe was dirt poor following the wars, a lot of people saw and caused a lot of social change during their lifetime.

Anus Mirabilis

"Convince us that they are dangerous or you will be fined"

The chinese don't view AI as metaphysical, they view it as an engineering challenge they consider good for their state and want to dominate the global market like they do with batteries/EVs/photovoltaics. They want to proliferate them as much as possible and traditionally they don't care much for IP. They also want hardware makers to make optimized chips specifically for these models.

And that is proof that it has arrived; you no longer need to read anymore.

> Compute, energy, physical infrastructure, land, raw materials, and the speed at which new capacity can be constructed may become some of the principal bottlenecks of the post-AGI economy.

We 've been here before. This post-AGI era is a lot like the era of slavery. So what matters is basically the ownership of land (because you can use it to generate energy/compute, raw materials and build infrastructure on it). Bill Gates was right to invest in land.

Hail to the new feudalism!

The ISS has only existed in times of peace

Aside from practical issues, why make your datacenters so vulnerable?

Enrico Fermi will stop spinning in his grave

Still waiting for Linear A

the most intentional internet was made by people, not algorithms

now that we have cheap spam filtering, why don't we go back to forums

Imagine the parthenon marbles. When they were looted it was even a celebrated act, but they are still stolen in the british museum centuries later.

So now we have found a way to use brain organoids by grafting and growing them inside mouse heads. Those Frankenmice will be used as a platform from now on.

I m looking forward to talk to the first language-capable Frankenmouse.

It isn't going to stop until it becomes a national security issue. If your can spy on your citizens , your enemies can potentially gain access to the same surveillance and hit high profile targets.

We live in war times and will soon see it happening

They should be called speakeasys

in which country?

I believe soon we will need to instill religion into AI , leading to the real clash of civilizations, embodied by the frontier language models of (post)-christianity, islam, judaism, buddhism etc. Religion is language, after all

Is he trying to find a cop-out for his diminishing returns?

Hey what about Flutter

Surely the next version will unfold twice to reach A3 size

Humans do tend to consume a very large part of the resources of the planet. Surely a superior being would be doing some pest control in its planet, just like we exterminate roaches

Including them?

> the target function frontier labs are trying to approximate is ultimately a super intelligence

So an unknown function that is even unintelligible to humans? that sounds a nonstarter

> Neural nets create models and deep layers of abstraction around

We don't have any proof of that, but the approximator thing is proven. The rest is just marketing speak.

Synthetic data derive from other linguistic data. Whatever intelligence is in there, it is expanded horizontally, not vertically

What is the rationale for superhuman intelligence? Neural networks are approximators being fed human intellect. Therefore they can only approximate the intelligence of humans. Even if the llm speaks an alien language, it should be similar to human intellect. Moving to the vertical axis would require some different mechanism.

They should also have asked them if they believe corruption is bad.

The old PRISM servers got overloaded

The csuite has the power to keep AI away from it or make itself mandatory.