31–40 of 228 posts
Somewhat counter quote... "Don’t worry about people stealing your ideas. If your ideas are any good, you’ll have to ram them down people’s throats." -- Howard Aiken ...to mean that, usually, the good ideas are the crazy sounding ones...
If its a good idea that's obvious, it's already used widely. If its not obvious, you'll still have to convince people. None of that requires lots of lies, though.
That implies that it is ridiculously easy to be right when everyone else is wrong. People aren't trying to be right. Any sort of principle-based analysis easily outperforms the herd. When leaders in society start lying that is indeed one of those situations. Pretty much any situation where everyone knows something and the hard statistics are telling a different story is.
The more pressing problem is how to go from a lovable Cassandra to someone who can preempt major events and convince the herd to not hurt itself in its confusion. Coincidentally that is how markets work, people who have a habit of being right are given full powers to overrule the mob and just do what they want. Markets don't care if everyone believes something. They care if people who got the calls right last time believe something.
In this case, the US hasn't seen a good outcome to a war since something like WWII and even there they waited until the war was mostly over and the major participants in the European theatres were exhausted before getting involved. The record is pretty bad. Iraq was an easy call to anyone who cares about making accurate predictions.
Makes me think of academic papers that overhype their contribution. Also makes me think about AI hype.
For me the danger of AI is that it enables the surveillance state through facial recognition and the instantaneous aggregation of all my data. For "national security" reasons, I may be detained and denied of my rights if Palantir hallucinates. Who do I sue if Palantir decides I am an illegal?
Earlier quoted context omitted.
For me the danger of AI is that it enables the surveillance state through facial recognition and the instantaneous aggregation of all my data. For "national security" reasons, I may be detained and denied of my rights if Palantir hallucinates. Who do I sue if Palantir decides I am an illegal?
Ot worse because it didn't hallucinate, and they are coming for you, as a free thinking "radical". They can tell from a long deleted blog post you made in 2005 about green energy.
Put it this way, if Hitler had grok, would it really get any worse for the Jews? I don't think so. I think they would be screwed no matter what.
It is also a useful trick to keep in mind the opposite of critical thinking - following the herd. Just copying everyone around you is often a great strategy. So good that even if everyone around you are making mistakes it can still be the dominant strategy (there is a reason a lot of people who don't like war are cowed into silence when war fever descends). Most people are using it. That implies that it is ridiculous…
I think this true but misleading — conditioned on other people going with the herd in the wrong direction, it is easy to be right. However, often the herd is going in a right (or at least acceptable) direction. The continual effort to check if the herd is going in the right direction _is not_ easy. If a magic eight ball could alert you “hey the herd is wrong right now, take a closer look”, that would be great! But we have no such magic eight ball.
Having worked in public advocacy advertising, I’d frame it like this: “Good ideas don’t need lies” is a compelling ideal but in practice, public acceptance isn’t a reliable signal of truth or societal benefit. It depends on incentives, narratives, and how information is presented. History shows that even harmful or suboptimal ideas (like coal power) can gain widespread support if presented persuasively, while genuine…
I think you just reinforced the articles point. Coal power needs lots of lies to justify it, as per your own statement. That is in fact because coal energy is a terrible idea . It has 0 upsides compared to renewable alternatives, and is on the whole worse than even other non-renewable alternatives. If you have to lie to make it sound good, that's probably because it isn't actually good
My point is just that public acceptance itself isn’t reliable evidence either way: ideas can gain support (or fail) for reasons other than their actual merit.
Earlier quoted context omitted.
Ot worse because it didn't hallucinate, and they are coming for you, as a free thinking "radical". They can tell from a long deleted blog post you made in 2005 about green energy.
Why bother with all that though? Just ask them to do their job for the party. If they don't, or you suspect they don't align with the party, you just execute them. Don't need tech for this. The tech is just for some people to get rich, not to really enable any new evil that can't already be achieved today with pen and paper and bullet (as modeled extensively in the last century). Put it this way, if Hitler had grok,…
Having worked in public advocacy advertising, I’d frame it like this: “Good ideas don’t need lies” is a compelling ideal but in practice, public acceptance isn’t a reliable signal of truth or societal benefit. It depends on incentives, narratives, and how information is presented. History shows that even harmful or suboptimal ideas (like coal power) can gain widespread support if presented persuasively, while genuine…
[dead]
Having worked in public advocacy advertising, I’d frame it like this: “Good ideas don’t need lies” is a compelling ideal but in practice, public acceptance isn’t a reliable signal of truth or societal benefit. It depends on incentives, narratives, and how information is presented. History shows that even harmful or suboptimal ideas (like coal power) can gain widespread support if presented persuasively, while genuine…
You just raised another example of a bad idea that needed lies to gain public acceptance.
I’m pointing out that the process by which ideas gain acceptance is somewhat independent from their actual quality, so acceptance alone isn’t a strong signal.
Earlier quoted context omitted.
> stock options did become nearly universal in tech compensation Although I've noticed that options have been replaced more and more these days with RSU's (plain old grants) because options have a tendency to go "underwater", suggesting that they weren't all that great to begin with.
It’s been standard advice on this forum for at least 10 years to value options at $0, and only consider cash comp + RSUs.