RandomRanger
Just build nuclear plants!
No bio...
User ID: 317
RandomRanger
Just build nuclear plants!
4d ago
So long as these beings have a conscious experience of their own and they are still capable of loving, fighting, fearing, suffering, and failing
Why would they have love? Why do we have love? In large part because we have sexual reproduction. Even without that there's parental love...
What is love for a machine that can copy itself out as necessary, self-modify directly? Can you feel love when it's just different versions of you (different parts of you?) or is it narcissism? We don't really have words for this.
Why would they suffer? Pain acts as a reinforcement signal, a warning. But if the full focus of the mind is already fixated on a goal then suffering is just wasted energy.
I think people underrate how alien a Powerful Being would be. Not merely some superintelligence derived from a mostly-human text corpus (already unprecedented and alien) but something that evolves or self-modifies into existence later on in a post-singularitarian environment.
This would be a fundamentally different environment to all organic life. Conscious self-analysis and self-modification is an absolute gamechanger, nevermind every other trump card these beings would have. I'd expect something fundamentally different to emerge.
RandomRanger
Just build nuclear plants!
5d ago
Who says HBD means all countries run by smart people are more pleasant than countries run by stupid people? I'd rather live in Fiji than North Korea. But Koreans are more capable than Fijians.
Also now I look at it, Russia has twice Mexico's GDP per capita PPP. And even if you look at median incomes, Russia is still about 70% higher. So I don't think that Russia is even poorer than Mexico in any meaningful way and murder rate is lower too. Hard to be more murderous than Mexico. So unless we look at nominal GDP figures which are distorted by sanctions, then Russia is clearly better off. HBD bros win again.
RandomRanger
Just build nuclear plants!
5d ago
They did do a tonne of buybacks, that's why their share price is so high!
Google, 317 billion of buybacks in 10 years. Meta $208 B in 10 years. Apple $735 B... Over 1 trillion in 5 years is nothing worth sneezing at.
They have since lowered buybacks because of AI spending. Trillions in share buybacks absolutely does drive stock prices.
RandomRanger
Just build nuclear plants!
5d ago
I was just trying to be a little artsy and alliterative... Don't read so much into it.
RandomRanger
Just build nuclear plants!
5d ago
Good point, that's what I meant but not what I said!
Russia isn't a counterexample against HBD. The USA was reliant on Russian spacecraft to reach the ISS for some time (which again the Russians helped build). Likewise the US relies on Russia for about 20% of its enriched uranium.
Mexico doesn't produce machine tools, oil industry equipment, aircraft engines, rocket engines, radar or any sophisticated technology at all. Mexico only assembles electronics. They assemble cars and components designed by others, with machinery supplied by real industrial powers.
Russians are far more capable than Mexicans, which is why the US casually dominated Mexico whenever they clashed, whereas Russia was and is much more of a challenge. Russia can reduce Europe and North America to ash in hours, Mexico can barely control its own territory from gangsters nevermind prosecute foreign policy.
RandomRanger
Just build nuclear plants!
6d ago
There are tradeoffs.
Slutty Sabrina might want Chad Centimillionaire and settle for Nicholas Niceguy. Patrick the Patriarch might want Virginal Virginia and settle for Somewhat Slutty Sarah...
Also 25% of US adults under 40 have never married. So it is completely plausible that Somewhat Slutty Sarah is just not going to marry, same for Nicholas and Patrick for that matter. 25% of the US population seems not to be marriage material.
Have you missed the last 30-40 years of financialized capitalism where shareholders encouraged companies to forego investment, cut R&D and provide short term profits?
Aggressive investment has absolutely not been rewarded by shareholders in the recent past. And nor is it rewarded today.
The company reported first quarter 2026 earnings results on Wednesday and raised its full-year 2026 capital expenditure guidance to $125 billion to $145 billion, up from a previous range of $115 billion to $135 billion. Meta told investors the boost was the result of higher prices for components and “additional data center costs to support future-year capacity.”
Last year, Meta spent $72.2 billion on capex, up roughly $30 billion from the year before. The company is now guiding to nearly double what it spent in 2025, and more than it spent in 2025 and 2024 combined.
In after-hours trading, the stock tumbled more than 6% as a result of the jump in capex guidance. In contrast, Alphabet and Amazon—which are also spending enormous sums on AI infrastructure buildout, and which both announced earnings on Wednesday—saw their share prices rise after hours, in part because they both reported AI-related growth in their massive cloud-services businesses.
Investors understandably are not rewarding companies just for investing in AI without showing that it's raising revenue (which it is, even with Meta who's making a fortune off improved ad targeting). They only reward investment that seems to be delivering returns. And even then they're skittish about big investments.
I've made one such auto-proofreader, I use it, it works just fine bringing up real mistakes for me to fix, hardcoded what kind of English I'm using, only a handful of false positives that I can quickly ignore. Every white collar worker I've shown it to says 'huh this is really cool, picks up things I've missed, saves lots of time' and they routinely ask to use it...
Absolutely massive own goal by Microsoft and big tech that most people's main experience of AI is a derpy chatbot that mangles documents or genericises things and not well-designed processes to solve specific tedious problems, nor the extremely flexible coding tool I used to make the auto-proofreader.
RandomRanger
Just build nuclear plants!
6d ago
Superintelligence by end of 2027 is roughly what Anthropic/Dario seem to predict, so it's only roughly as maximalist as the most bullish frontier lab. The AI 2027 crowd backed out a year or two but I think they were roughly on the money the first time in their analysis. I'm not so keen on their 'centralize all US AI research and hand ultimate authority to a chosen council of experts' proposal though.
Let's suppose AI models aren't so great for mathematics and it's Terence Tao doing most of the work.
The primary commercial usecase for AI isn't mathematics, it's coding, as you say.
I cannot code beyond the most basic Khan Academy beginner sense, my actual end to end abilities are completely worthless. And yet I can make useful AI tools for work, processing documents in various ways that saves lots of human time. In a certain sense, I'm providing most of the 'secret ingredient' since you cannot just tell an AI to do these fiddly tasks and expect them done properly in oneshot. It will usually not work the first time. So I give it some counsel and tell it what to troubleshoot, errors, differences from expected output, clarify my intentions and ultimate usefulness. Eventually it works and then I refine it to work better and better by getting AI to handle all these edgecases and Word-induced BS.
And how could I make a (still under development) 4X game with AI if I can't code? There's a fair bit going on. Space battles, ground battles, culture, technologies, buildings, resources, goods, markets, map generation, turn order, trade between provinces (intra faction), trade between factions, freighters, pops and social classes, loans, diplomacy and war plotting, coalition building... Some things are not well fleshed out but there is quite a bit there.
I was just now getting it to make an evolutionary testing system to refine ship designs and fleet compositions and so define the meta. First time it worked OK, then when trying to make it better (too many bad mutations!) it broke, then I overhauled it and now it works great and with multicore processing too. Apparently the dominant strategy is getting hundreds of incredibly cheap and terrible warships to act as chaff for a small core of high-tech warships to exploit the targeting and reinforcement logic. So clearly I need to change how reinforcement and targeting works, raise minimum costs for ships.
It was my idea to make this tool, my idea behind the overhaul and my ideas behind every mechanic but I could never have done it myself. The secret ingredient is clearly the AI.
More options
Context Copy link