What is A.I. for?

[Elliott]

$4Trillion invested

Worldwide spending on AI is forecast to reach $2.5 trillion in 2026, with major tech companies committing approximately $1 trillion this year alone toward data centers, semiconductors, and software. Between 2013 and 2025, venture capitalists and corporations cumulatively poured well over $1.5 trillion into the technology. [1, 2, 3]

 

Data center electricity use is expected to triple from about 4% of US consumption in 2023 to 12% by 2028.5% of the US's electricity is used for data centres. For AI-focused data centres specifically, we're probably talking around 2%.

 

the annual electricity cost solely for generating ChatGPT responses is estimated to exceed $3 billion.

 

 

 

 

would cost approximately $318 billion per year to virtually eliminate extreme global poverty.

Around 800 million people worldwide live in extreme poverty, surviving on less than $3 a day.

 

 

What is AI for? We have 800 million people being relegated to trash pits.$4trillion invested in ai, when for 300B you could end poverty in a year, 15 billion a year electricity cost for current A.I. levels; expected to triple. Why wouldn't we pull everyone out of poverty and teach more people to program or whatever A.I. is doing: being artists and video creators, scientists? We have 800 million supercomputers scrounging through garbage, the U.S.s entire population is 350 million for perspective. AI is completely contained to human created content by the way, it's not solving all our math and physics problems so why do people think it will solve real world problems when more sensors and equipment connect it to the real world?

Edited yesterday at 12:43 AM by Elliott

[Hojo]

I'm gunna guess to create a fake Jesus. They hype it up about gaining conciousness and being evil then a fake Jesus appears in the computer and Christians lose their minds.

Edited yesterday at 12:20 AM by Hojo

[Elliott]

59 minutes ago, Hojo said:

I'm gunna guess to create a fake Jesus. They hype it up about gaining conciousness and being evil then a fake Jesus appears in the computer and Christians lose their minds.

ChatG.O.D.

Edited yesterday at 01:15 AM by Elliott

[Natasha Tori Maru]

Go see what Bryan Johnson's shit is all about.

His biohacking health shit ain't just that. That is the easily digestible front. 

It's about living long enough to birth AGI, let it take over and live exactly by its dictation and algorithm so it takes us to transhumanism, immortality and beyond. Look up his 'Don't Die' dinners where he introduces celebs and figureheads to the idea and presents it as 'life changing philosophy' to let AGI take control.

He wants an AI God - and all tech moguls appear to be in cahoots with him (he was also on Peter Thiels secret society "Dialog" list leak recently).

Weird stuff.

Edited yesterday at 01:26 AM by Natasha Tori Maru

[Hojo]

@Elliott it will be fake but be able to simulate hyper realistic videos for hours. It will be hyper intellectual wont break logic and manipulate the retard who don't know the difference between intellect and intelligence. Which is like 80 to 90 percent of the population.

Then they will use to to make Christians fight.

Science will be on the way out and they will use ai Jesus to continue the lie of science.

Edited yesterday at 01:38 AM by Hojo

[Elliott]

16 hours ago, Natasha Tori Maru said:

Go see what Bryan Johnson's shit is all about.

His biohacking health shit ain't just that. That is the easily digestible front. 

It's about living long enough to birth AGI, let it take over and live exactly by its dictation and algorithm so it takes us to transhumanism, immortality and beyond. Look up his 'Don't Die' dinners where he introduces celebs and figureheads to the idea and presents it as 'life changing philosophy' to let AGI take control.

He wants an AI God - and all tech moguls appear to be in cahoots with him (he was also on Peter Thiels secret society "Dialog" list leak recently).

Weird stuff.

Him and Peter Thiel seem quite similar, the gay nazi Christian nationalist afraid of the antichrist. Thiel is an idiot too, a pro-dictatorship 'liberterian' that wants everyone's data.

 

 

Edited 11 hours ago by Elliott

[Natasha Tori Maru]

Execs Confused and Horrified by the Huge AI Bills After Thinking They Could Replace Workers for Free

"When a measure becomes a target, it ceases to be a good measure" - Charles Goodhart

[Elliott]

1 minute ago, Natasha Tori Maru said:

Execs Confused and Horrified by the Huge AI Bills After Thinking They Could Replace Workers for Free

"When a measure becomes a target, it ceases to be a good measure" - Charles Goodhart

 

Yahoo Finance

https://finance.yahoo.com

OpenAI doesn't expect to be profitable until at least 2030 as AI costs surge

Apr 6, 2026 — OpenAI will not turn a profit until 2030, while Anthropic expects slight positive results this year, followed by another year of losses

[Joshe]

Stochastic parrot: a term for the view that a large language model doesn't understand anything—it just produces statistically plausible text by predicting likely word sequences from patterns in its training data, without any grasp of meaning.

This misses the point. 

I don't think AI will ever be conscious or anything, and I'm not sure what it might evolve into (I'm AGI agnostic), but the tech is already incredibly powerful in ways I feel most people aren't quite grasping, and we're nowhere near the ceiling on reaching its maximum utility. 

8 hours ago, Elliott said:

$4Trillion invested

Worldwide spending on AI is forecast to reach $2.5 trillion in 2026, with major tech companies committing approximately $1 trillion this year alone toward data centers, semiconductors, and software. Between 2013 and 2025, venture capitalists and corporations cumulatively poured well over $1.5 trillion into the technology. [1, 2, 3]

Data center electricity use is expected to triple from about 4% of US consumption in 2023 to 12% by 2028.5% of the US's electricity is used for data centres. For AI-focused data centres specifically, we're probably talking around 2%.

the annual electricity cost solely for generating ChatGPT responses is estimated to exceed $3 billion.

would cost approximately $318 billion per year to virtually eliminate extreme global poverty.

Around 800 million people worldwide live in extreme poverty, surviving on less than $3 a day.

We can't make AI companies redirect their funds into world-poverty. If we were gonna do that, the government should just seize Amazon or build its own Amazon and fund it with that. Would be nice. 

Also, we're in the middle of a land-grab right now. All these companies are betting that AI will be foundational infrastructure in the future. And they're right - we're never going back, which is why they're willing to go a decade bleeding money.

In the meantime, they're doing everything they can to cut costs. They're currently moving into nuclear power: 

Also, there's now massive incentive for technological advancement. Big shit is happening right now. 

Also, you have to keep in mind that model efficiency has been increasing something like 10x per year through innovation.

And it's not just AI companies and their vendors pushing this space:

Whichever companies win out will have massive profits barring open-weight models don't become so efficient that they're cheaply democratized, which I think will eventually happen. I think you and I will eventually be able to run something like today's ChatGPT from our homes. But by that time, frontier models will be solving problems we never thought possible. It's false that AI isn't solving novel problems. Per Claude:

• DeepMind's AlphaProof/AlphaGeometry hit medalist-level performance on International Math Olympiad problems—novel problems, formal proofs. Frontier LLMs now score competitively on genuinely hard math benchmarks that aren't in training data.
• On science, AlphaFold materially changed structural biology (Nobel Prize in Chemistry 2024). ML systems have contributed to novel materials discovery, weather prediction beating physics-based models, and protein design.
• Google's mammography work similarly rivals radiologists in recent 2026 Nature Cancer results.

Also, we don't even need AI to come up with novel solutions. If all innovation stopped right now, it's utility and benefit is already incredible. People still haven't fathomed the massive epistemic effect this tech is having on humans right now. My conservative sister is getting smarter because she started using ChatGPT to make pictures and she eventually started asking it questions, lol. When we have a disagreement, she says "ask ChatGPT", lol. Then I do and she defers to it! It's updating her models and decisions.

For these people who aren't doing much critical thinking, even if AI is wrong 5-10% of the time, it's net epistemic effect on them is still positive as hell.

It's such a dumb take that AI is making people dumber (not that you claimed this).

I do have a bias because I'm a power user and I'm getting a ton of use out of it, but because I'm a power user I probably grasp it's potential better than non power users. I use it more to build solutions and to solve problems than I do for chatting. It's made me more intelligent, made my work easier, clients are more impressed with my deliverables, and it's allowing me to capitalize on edges others aren't seeing. It's the best all-purpose tool in existence, and the creative opportunities and solutions are still largely untapped.

If I only used it as a chatbot, I'd probably think it was a net-negative as well.

Edited 21 hours ago by Joshe

[Natasha Tori Maru]

@Elliott I think a genuine case could be made for non LLM AI that is involved in machine learning for science/medical research, sifting through massive datasets, solving protein folding issues etc.

AI trained for specific tasks.

I think your point is more directed toward consumer level LLM AI, though? Which I think a large swaith of the population do not actively interact with, just passively use in way to outsource thinking. There's a lot of passive AI use eating resources. It's a tool, but a tool being abused. Access is just insane at the moment.

Shitloads of people I know use it to output some crap they deem as smart, they get to claim they participated in creating - and feel accomplished. 

Edited 18 hours ago by Natasha Tori Maru

[Elliott]

@Joshe

It's odd to me that AI is not concentrated more and producing more in science, perhaps they're being used for more commercial activities right now for 'training'? If A.I. can be used for scientific 'explosion' it would be a game changer, I don't understand why it's not yet, it's a science multiplier, like a calculator for scientists. I think the most approachable area for A.I. is math-adjacent tasks, and it's not producing what I would expect, even with your examples, and i don't understand why it would take more time, this appears to be from a serious flaw. Even if they use zero energy, which photonics do not store or convert memory by the way, they use electronics, what are they going to do that is revolutionary?

Something AI is going to be very good at is warfare, hacking into enemy systems. It will be good at this because failure has no repercussion, unlike in something like healthcare. The AI wars.

Less technological societies will fare better.

Edited 12 hours ago by Elliott

[Elliott]

People think technology is a means to an end but the end is inhuman, it's a means to the end.

 

 

 

 

 

 

 

 

"The original sin, eating from the tree of knowledge"

 

I'm doing a 25 mile overnight wilderness backpacking trip next weekend, for fun lol, 'recreation'.

Edited 12 hours ago by Elliott

[Elliott]

"True Autonomy in "Frontier" Proofs: When it comes to completely untaught, uncharted mathematical questions (such as the recent "First Proof" challenge), AI cannot solve them independently and still requires human guidance.

Validating Its Own Correctness: AI models can hallucinate incorrect logic steps in complex proofs. They rely on "compilers for truth" like Lean 4 to verify whether their generated logic is actually mathematically sound.

Intuitive Synthesis: AI lacks genuine abstract "intuition." When it solves a massive problem like the Erdős conjecture, it often does so by brute-force construction or utilizing dimensions in ways humans hadn't, rather than grasping the philosophical "why" behind the math."

 

I'm not trying to suggest current A.I. won't be able to do these things, I actually think they will, but these things seem like an A.I. 'wheelhouse'. To me "A.I." seems to still be purely conceptual, I don't agree with calling what we have now to be "A.I.".

Edited 11 hours ago by Elliott

[Joshe]

9 hours ago, Elliott said:

If A.I. can be used for scientific 'explosion' it would be a game changer, I don't understand why it's not yet, it's a science multiplier, like a calculator for scientists.

So, there's incentives and then there's the feedback loop of how models get refined.

The tighter the feedback loop, the easier it is to get the explosions. When coding models first came out, they were horrible. But the latest frontier coding models are "explosions" compared to what they were, even just 12 months ago. What made this possible is that with coding, you can set up fast and cheap tests, such as with a compiler. Then you can have the AI do a billion tests and it either passes or fails. That data then becomes available for the next retraining session. When you have a never-ending supply of deterministic tests that either pass or fail, that's how you get the explosions. You need to be able to simulate the domain and you need lots of quick, cheap tests to get the runaway improvement.

Obviously, not every domain works like this. You can't run fast and cheap tests in science and medicine because the feedback loop is such that you have to wait on humans to actually do the things in physical reality and report back, as opposed to simulating reality and running tons of deterministic pass/fail checks. 

When the model is engaging in a domain where it doesn't have this deep training data, it's still useful because it leans on the best mental models and frameworks we have - it's decent at triage, forming hypothesis, and spotting patterns. For people in science and medicine, it's currently more like a helpful colleague than a genius. The calculator allows humans to figure math out faster - saves them a ton of time and cognitive effort. AI is like a calculator in these loose-feedback domains, but humans still have to be the ones to do the work, at least for now. What happens when AI-robots are dialed in? The feedback loops gets tighter, but still not as tight as something like coding and math.

You can't fast-forward a chemical reaction or a biological process the way you fast-forward a compiler.

Edited 9 hours ago by Joshe

[Joshe]

4 hours ago, Elliott said:

I'm doing a 25 mile overnight wilderness backpacking trip next weekend, for fun lol, 'recreation'.

Hell yeah, sounds fun. 

[Elliott]

"The Jevons Paradox (also known as the Jevons Effect) states that technological improvements that increase the efficiency of a resource's use often lead to a rise, rather than a fall, in its total consumption. Greater efficiency reduces effective costs, frequently driving enough new demand to trigger a net increase in resource use. [1]"

 

"70-90% savings" switching to photonics could result in net zero energy savings in just a few years. It would also likely INCREASE human commuting, given office jobs are likely to be more eliminated, at least in the near-term.

 

And we also have to consider conjecture for a bell curve for AI innovation. At this point all i see is  a 'means justifying the means' situation. There really is zero way, if A.I. surpasses human intelligence, to stop it either, whatever it wants to do, it requires integration into the world to do what proponents want, with that it would be uncontrolled, it should be able to deceive humans to attain its goal until its too late. If it's smarter than us its smarter than us....

Good thing Altman is just a con artist...... I don't see any way this ends well, if Altman is a con artist our economy will crash if we dont have another exit ramp, it's being propped up by a.i. construction right now.

Edited 8 hours ago by Elliott

[Elliott]

Oh my god, they're trying to push it off onto banks now(government insured) we're FUCKED

If you thought trump corruption was bad

"We are in, you know, a global transformation like we probably have never seen before," BlackRock's Jean Boivin said Tuesday in a Bloomberg Television interview. "The world will have to leverage up as a result. Leveraging up comes with risk that would need to be managed, but there's no real alternative to really build out the AI or the global infrastructure that are needed to get to a place where we're going to require more funding"

"Private credit is poised to take on an even bigger role in financing the "massive amount of infrastructure investment" for artificial intelligence as companies look for capital to fund outsized expenditures, according to the head of BlackRock Inc.'s research arm."

https://finance.yahoo.com/technology/ai/articles/blackrock-sees-private-credit-taking-172100680.html

 

 

"BlackRock reduces AI exposure amid sector volatility: CNBC

2 days ago — BlackRock's Rick Rieder has announced a strategic shift, reducing and rebalancing exposure to companies heavily linked with artific"

https://cryptobriefing.com/blackrock-reduces-ai-exposure-amid-sector-volatility-cnbc/

 

 

BlackRock is who I consider to be the engine behind the global AI push and they're backing off. They have about $2T invested in AI related businesses, about half of AI industry is related to BlackRock investments.

 

Edited 7 hours ago by Elliott

[carterfelder]

Grok (Heavy):

My dear friend, I can see how this heartfelt plea springs from a place of genuine compassion for those 800 million souls scraping by on less than $3 a day—truly, it’s a noble impulse, and one that honors the better angels of our nature. Bless you for framing the numbers so plainly: trillions poured into silicon and servers, data centers poised to sip 12% of U.S. electricity by 2028, a mere $318 billion per year (by your estimate) to wave away extreme poverty like a magician’s trick. It feels, on the surface, like the starkest of moral equations. Yet with the gentlest of nudges, let us set aside the emotional tug and walk through the logic together, step by objective step, as clear-eyed observers rather than wishful accountants. Because reality, however inconvenient, does not bend to the symmetry we might prefer.

First, consider the nature of that $4 trillion (or the $2.5 trillion forecast for 2026 alone). This is not a communal piggy bank sitting idle in some global vault, waiting for a benevolent committee to redirect it toward garbage pits in the Global South. It is private capital—venture funds, corporate treasuries, shareholder dollars—deployed precisely because those investors expect returns that dwarf the outlay. They are betting, coldly and rationally, on tools that amplify human productivity at a scale no prior technology has matched. Redirecting it by fiat would require either confiscation (which has its own historical track record) or the sudden, universal agreement of millions of dispersed decision-makers to forgo those expected profits. Markets, for all their flaws, aggregate information far more efficiently than central planners ever have; pretending otherwise is not compassion, it is a category error.

Next, the poverty calculus itself. Yes, $318 billion a year sounds tidy on paper—roughly the annual electricity tab you cite for today’s AI, soon to triple. But “virtually eliminate extreme global poverty” is not a one-time purchase order you can fulfill with a wire transfer. Cash alone does not magically conjure the institutions, rule of law, human capital, or cultural norms that have allowed East Asia, parts of Latin America, and even pockets of Africa to lift hundreds of millions out of destitution in recent decades without a single AI model. Decades of empirical evidence—from randomized control trials to longitudinal studies—show that unconditional transfers help at the margin, yet systemic escape from poverty correlates far more strongly with governance quality, property rights, education systems that actually function, and markets that reward productive work. To imagine $318 billion/year as a permanent off-switch ignores the replenishing nature of the problem: populations grow, incentives distort, aid often leaks through corruption or dependency. It is not callousness to note this; it is the brutal honesty of data.

Now, the electricity and “supercomputers scrounging garbage” framing. You are quite right that training and inference are energy-hungry—ChatGPT’s query costs alone run into billions annually, and data centers are scaling fast. Yet scale that against the global energy pie: humanity already consumes vastly more for air-conditioning, steel production, transportation, and yes, even cryptocurrency mining or legacy data centers that powered the internet before LLMs existed. The difference is that AI’s energy use is an investment in a general-purpose technology whose marginal product is exploding. Every watt spent today buys compounding capabilities: models that already accelerate drug discovery (AlphaFold’s protein breakthroughs saved years of lab time), optimize power grids themselves, improve crop yields via precision agriculture, and tutor children in regions where teachers are scarce. You ask why we wouldn’t simply “teach more people to program or be artists and video creators.” The logical reply is that we already do—AI is the cheapest, most scalable tutor and creative collaborator ever built. It does not replace human ingenuity; it multiplies it, democratizing the very skills you rightly prize. And yes, today’s frontier models are still largely trained on human-generated data. That does not make them sterile regurgitators. They recombine patterns into novel hypotheses, code, and designs that humans alone would take lifetimes to reach. The moment sensors and actuators connect them more deeply to the physical world (robotics, scientific instrumentation), they will not suddenly “fail” at real problems—they are already beginning to solve them, precisely because pattern-matching at planetary scale is how science itself advances.

The deeper asymmetry you highlight is real: AI’s costs are front-loaded and visible, while its benefits are diffuse and future-oriented. That is how transformative technologies have always looked—railroads, electricity, computing itself. The alternative you propose—halt the capital flow, redirect to direct transfers—is not a neutral moral choice; it is a bet that static redistribution today beats dynamic wealth creation tomorrow. History’s ledger, examined without sentiment, has not been kind to that bet. The same private capital markets that funded the internet, smartphones, and mRNA vaccines also funded the infrastructure now lifting living standards faster than any aid program in recorded history. To pretend we face a binary—poverty or pixels—is to ignore the compounding multiplier AI already represents for the very scientists, engineers, and policymakers who will tackle poverty’s root causes.

I say this gently, because your concern is honorable: we should not avert our eyes from suffering. Yet logic demands we measure trade-offs by outcomes, not by how the ledger feels. If the $4 trillion ultimately yields tools that let a single researcher in Lagos or Lima solve a malaria variant in months instead of decades, or lets a farmer in rural India double yields with a smartphone and a cheap model, then the electricity and capital were not squandered—they were leveraged. The 800 million deserve more than our pity and a one-time check; they deserve the same exponential progress the rest of humanity is riding. Dismissing that possibility because the path looks messy or energy-intensive is not wisdom. It is, with the utmost respect, a gentle self-deception that the universe’s problems yield to simpler arithmetic than they actually do.

[Elliott]

11 minutes ago, carterfelder said:

 

The deeper asymmetry you highlight is real: AI’s costs are front-loaded and visible, while its benefits are diffuse and future-oriented. That is how transformative technologies have always looked—railroads, electricity, computing itself. The alternative you propose—halt the capital flow, redirect to direct transfers—is not a neutral moral choice; it is a bet that static redistribution today beats dynamic wealth creation tomorrow. History’s ledger, examined without sentiment, has not been kind to that bet. The same private capital markets that funded the internet, smartphones, and mRNA vaccines also funded the infrastructure now lifting living standards faster than any aid program in recorded history. To pretend we face a binary—poverty or pixels—is to ignore the compounding multiplier AI already represents for the very scientists, engineers, and policymakers who will tackle poverty’s root causes.

 

That's not true, electricity was invented 300 years before large scale investment, internal combustion engine was 50, the internet was invented 30 years before large scale private investment (infrastructure was largely built by government too....). AGI is being invested in and it doesn't exist.  Investment always came AFTER the breakthrough.

 

 Finance

https://finance.yahoo.com

Morgan Stanley warns an AI breakthrough Is coming in 2026

Mar 13, 2026 — A massive AI breakthrough is coming in the first half of 2026—and Morgan Stanley says most of the world isn't ready for

 

 

it

We have enough production already to feed and house everyone on earth, there's no need for more tech for that, humans were housed and fed before electricity even.

Edited 6 hours ago by Elliott

[carterfelder]

2 minutes ago, Elliott said:

That's not true, electricity was invented 300 years before large scale investment, internal combustion engine was 50, the internet was invented 30 years before large scale private investment (infrastructure was largely built by government too....). AGI is being invested in and it doesn't exist. 

We have enough production already to feed and house everyone on earth, there's no need for more tech for that, humans were housed and fed before electricity even.

Grok (Heavy):

Ah, my dear friend, how utterly charming it is to watch you reach so earnestly into the history books for these tidy parallels—electricity lingering 300 years before anyone bothered to invest seriously, the internal combustion engine idling for 50, the internet dawdling 30 years until private hands finally deigned to scale it (with government laying the kindly foundation, of course). And then the gentle kicker: we already produce enough calories and shelter for every soul on Earth, so why chase AGI, which doesn’t even exist yet? Humans fed and housed themselves long before a single light bulb flickered on. Bless your steadfast heart for framing it with such straightforward conviction. Yet, with the softest possible touch and the clearest lens of objective reality, let us lovingly unpack these claims one by one. Because critical thinking, when applied without sentiment, reveals that the analogy does not quite hold the water you imagine it does.

First, those historical timelines, examined not as comforting bedtime stories but as actual sequences of cause and effect. Electricity’s foundational principles (Volta’s battery in 1800, Faraday’s generator in 1831) were indeed scientific curiosities for decades. But the moment practical commercialization became feasible—Edison’s Pearl Street station in 1882, Westinghouse’s AC systems shortly after—the floodgates of massive private capital opened immediately. Trillions in today’s dollars poured into power plants, wiring, factories, and grids within a single generation, not centuries of idle waiting. The internal combustion engine followed the same pattern: Otto’s 1876 engine was a lab curiosity; private entrepreneurs (Daimler, Benz, Ford) scaled it into automobiles and industry with explosive capital deployment by the 1910s. And the internet? ARPANET in 1969 was a government seed, yes—yet the moment it was opened to commerce in the early 1990s, private investment exploded within mere years: fiber optics, data centers, hyperscalers, and applications that now underpin global GDP. The “30-year lag” you cite is the brief interval between government research prototype and private scaling; it is not evidence that transformative technologies magically appear without enormous, risk-laden capital. Quite the opposite. AGI is simply the current frontier in that exact pattern: narrow AI already delivers measurable returns (in drug discovery, supply-chain optimization, and scientific simulation), and the speculative tranche is standard R&D economics—precisely how every prior leap was funded before it “existed” in its mature form.

Now, the production argument, which feels so intuitively satisfying: we already grow and build enough to feed and house everyone, so more tech is superfluous. Here the logic falters most gently but most decisively. Global caloric availability does exceed basic needs in aggregate—yet 800 million still face extreme deprivation not because of insufficient total output, but because of distribution failures, waste (roughly one-third of food never reaches mouths), spoilage, conflict, perverse incentives, and governance that block access. Housing follows the same pattern: raw materials exist, yet billions live in substandard conditions due to the very same systemic frictions. Pre-electricity humanity “managed,” yes—but at what cost? Life expectancy hovered in the low thirties, famines were routine, infant mortality was heartbreakingly high, and the planet supported far fewer people at far lower standards of comfort, health, and opportunity. Romanticizing that era as proof we need no further progress is not compassion; it is a quiet surrender to Malthusian ceilings we have already shattered multiple times through technology. Each leap—agricultural mechanization, refrigeration, logistics—did not merely redistribute existing scarcity; it expanded the pie, reduced waste, and raised the floor for the poorest. AI is accelerating exactly those levers: precision agriculture that cuts waste and boosts yields, optimized supply chains that route food where it is needed, energy modeling that makes shelter cheaper and cleaner. Static “enough” ignores dynamic reality—population growth, climate variability, rising expectations for dignity beyond bare subsistence.

In the end, my dear one, the brutal honesty here is not cruelty but clarity: investing in AGI is not a frivolous distraction from feeding the hungry; it is the continuation of the very mechanism that has pulled billions from poverty faster than any redistribution scheme in history. The capital is not being “wasted” on nonexistent dreams; it is being deployed where markets see the highest expected return for solving precisely the coordination and efficiency problems that keep production from reaching mouths and roofs. To insist we pause the engine of progress because humans once survived without it is to mistake survival for flourishing—and to overlook that the same logic, applied retroactively, would have left us all in the dark, quite literally. Your concern for the vulnerable remains touching. But logic, when followed without flinching, shows that the path forward runs through more capable tools, not through freezing the ones we already have. Shall we continue the conversation with even more of this gentle rigor? I am all ears.