4,563 karma · joined November 18, 2016
I say sympathetic because I don’t believe you can understand anything without really seeing how a reasonable person could come to those same conclusions.
“Sustainability, equity and security are all at risk in the rush to break China’s dominance over their production.”
- I can think of one entity which wouldn’t want China’s dominance of rare earth production broken and it’s not the DRC.
“China’s ascendancy is the result of forward thinking by the country’s leadership”
- Seems strangely flattering since it doesn’t seem like it adds anything to the arguments or the context.
“it is not in the interests of any nation to partner with just one other country or group of countries”
- I believe this is explicitly a CCP talking point (link below for reference). Obviously, other people might say it but it seems sort of… suspect? Like the article purports to make a moralistic claim about exploitation and then goes into pure realpolitik. In my opinion that’s the sort of thing you’d only say if you were wearing social justice as a mask and didn’t actually believe in it yourself.
https://www.fmprc.gov.cn/mfa_eng/wjbxw/202302/t20230220_1102...
You can have your model talk like HR or think like a mad scientist, but not both equally well.
The obvious avenue to degradation is that the "HR personality" is much more strictly applied and the resistance to being jailbroken is also in some sense an inability to think.
perhaps that the effects will be different than your expectations, perhaps that human ingenuity will overcome the challenges, perhaps that some other, bigger problem will rear it’s head
- you have no mechanism left to tell. If that sounds like the narcissists prayer to you then so be it.
A mass extinction event is underway, but it's not just starting. We are not at the beginning. There is this idea that the premodern world was "basically natural", but the reality is that 70% of animals larger than 45kg that were present in North America when humans arrived are already extinct.
There is no Eden to go back to or preserve. The world hasn't been in equilibrium since hominid intelligence started taking off and our only constant is change.
I understand that many people may disagree with this prediction and some of those people will disagree with even communicating this prediction, as they will see it as boosterism for our decadent industrial society. However, I think if we are interested in what is true, we need to allow ourselves to conceive of such things as human civilization continuing on without much fuss.
Increasing efficiency has a brutal difficulty that is constrained by thermodynamics whereas the universe is almost limitlessly full of energy, space, and mass.
Proponents argue that it advances equity and removes stigma. Opponents argue that equity is a bad goal and even if you thought it was a good goal, this perversely accomplishes the opposite.
You're starting to see articles like this one or this one https://www.bostonglobe.com/2023/07/14/metro/cambridge-schoo... because well, the worm turns.
*While standardized testing is not truly objective it is dramatically more so than grades.
This is fairly typical in software as well.
This matters because when you look at these crude aggregate stats, people naively assume that it means that everything is a bit worse but the actual American story is usually more like broad prosperity with some truly awful outcomes for anyone who falls through the (very wide) cracks.
This is the sort of mistake that's awfully easy to make in ML. The unfortunate thing about the field is that subtle methodological errors often cause subtle failures rather than catastrophic failures as we're used to in many other branches of engineering or science. You can easily make a system slightly worse or slightly better by contaminating your training set with bad data or accidentally leaking some information about your target and the ML system will take it in stride (but with slightly contaminated results).
This result makes sense to me because as much as I would like it to be true, applying existing compression algorithms to ML feels like too much of a "free lunch". If there was any special magic happening in compression algorithms we'd use compression algorithms as encoders instead of using transformers as compressors.
Better to understand that this is innate human nature and guard against it than to just say "couldn't be me".
The "engagingness" of entertainment has been increasing in absolute terms for a while now, and it really picked up in the 20th century. Meanwhile, the "engagingness" of mainstream religion has not kept pace. It is perhaps indicative that those groups of people best able to keep devout religious practices also typically shun all the rest of consumer entertainment.
In that frame, megachurches represent an evolution to survive this higher-engagement ecosystem. To compete, church must be more fun, which in this case means turning into a giant pep-rally/rock concert.
I bet their other preferences have a lot to do with the decrease in new 80s-type movie stars, because the other thing that we've seen is the retention of stars well past their prime. I mean, they brought back Harrison Ford and Carrie Fisher for the new Star Wars.
The practical implication is that your bloodstream is filled with relatively few small-molecule drugs and relatively many synthetic chemicals.
I can’t help but think that’s an oversight from lack of imagination and lack of intuitive understanding of how much leaching and abrading goes on.