We need to summarize the 4 most prevalent themes of opinions expressed in this Hacker News discussion. The discussion is about AI (artificial intelligence) and also references covid etc but the main discussion appears to be about AI's impact: de-skilling, usefulness, social effects, water usage, etc. But we need to identify the 4 most prevalent themes expressed by users.
Let's parse the discussion. The OP "righthand: Artificial Unintelligence" presumably is an article criticizing AI. Then many comments talk about various aspects: cycles of stupidity, covid, etc but the main focus seems to be AI's effect: de-skilling, overhyped, water usage, environmental impact, social damage, management shift, etc.
We need to find 4 most prevalent themes. We'll need to count recurring topics. Let's scan comments.
Topics seen:
-
Cycle of societal stupidity / repeating patterns (grebc: society loses collective mind every 10 years, war on terror, GFC, crypto, covid, AI). Also discussion about covid lockdowns etc, but maybe that's not core.
-
Critique of AI: de-skilling, making us lazy, losing analytical ability, AI as tool that reduces thinking, dependency.
-
Benefits of AI: productivity, code writing, debugging, learning, speeding up tasks, helpful for certain tasks, augmenting abilities, allowing focus on higher-level.
-
Environmental concerns: water usage of data centers, electricity, pollution, etc.
-
Social/damage: loss of community, social fabric, mental health, etc (covid discussion but also AI related? Some talk about social damage from covid, not AI.)
-
Management/agentic workflow: using AI agents, managing them, steering, etc.
-
Skepticism about hype: AI is overhyped, bubble, etc.
-
Concerns about power concentration, capital, monopolies.
-
Concerns about hallucinations, reliability.
-
Opinions about regulation, incentives.
We need to identify four most prevalent themes. Let's try to approximate counts.
We'll go through comments and tally themes.
I'll list each comment with themes.
Start:
righthand: OP (article) - presumably criticizing AI.
grebc: cycles of stupidity (theme: societal cycles), covid, AI.
doginasuit: sympathetic view of covid, vaccines, community resilience.
grebc: critique of lockdowns OTT, kids missing out.
watwut: adults fear vaccines.
fabioborellini: lack of human communication, cut friendships.
moron4hire: destruction of meetup groups, social damage.
kbelder: damage to children's education.
preg_match: lack of human connection trend, communities running on fumes.
dofm: community took six years to return, kids resent.
preg_match: resentment towards reality not handling.
dofm: kids angry about handling, specifics.
preg_match: lockdown reaction.
grebc: millions die in Australia? etc.
ChrisLTD: US less lockdown.
carbyau: US less lockdown, Australia low deaths.
teamonkey: Australia locked down hard, correct.
qsi: Melbourne longest lockdown.
carbyau: lockdown not equally spread, parties ignoring.
defrost: details of lockdowns.
simonbarker87: if works looks like overreaction.
doginasuit: hospitals overrun, need lockdowns.
carbyau: details of early COVID, unknowns, etc.
jakzurr: thank you.
pelotron: covid as dress rehearsal, US failed.
alexashka: journalists fear mongering, manufactured mass stupidity.
OutOfHere: disagrees, says concepts alive.
grebc: none inherent problems.
tenuousemphasis: TIL global pandemic not inherent.
MomsAVoxell: cycles every 4 years, materialist consumerism drives amnesia, etc.
andrekandre: (no text)
micromacrofoot: lost minds, LLMs produce good stuff, screwing up global economy, etc.
iAMkenough: both extremes true.
micromacrofoot: (re: god/fad)
Terr_: describes Enki.
zahlman: Enki definition.
Terr_: Enki data-center-adjacent.
iAMkenough: soon and god subjective.
alexashka: what is screwing up global economy? LLMs?
pelotron: concentration of capital into few tech companies, circular financing.
micromacrofoot: 25% of US GDP growth from AI money, crash.
S-E-P: vent post.
phyzome: perspicacious.
S-E-P: thanks.
daedrdev: AI does not use much water, water fear mongering propaganda.
S-E-P: datacenters stinky, ruin property values, pollute water.
scarmig: complaining about water usage indicates not well-informed.
S-E-P: need to grapple with merits.
scarg: no.
S-E-P: If someone spreading deception, point out without engaging further.
tptacek: CyrusOne data centers smell like grass, better than Walmart.
S-E-P: nicer area, Walmart etc.
AlexandrB: datacenter stinky? How? Pollute water? etc.
autoexec: Amazon data center polluted water source, nitrates, etc.
gorjusborg: EESI article: data center consumes up to 5 million gallons per day.
6031769: Consume how? staff drinking? kit not.
gorjusborg: 80% evaporates.
HedonicEscal8r: 450 million gallons a day for US data centers, equivalent to 90 towns.
gorjusborg: not asserting.
preg_match: sounds like lot but not much, almond requires 1000 gallons, beef etc. Go vegan negates AI water use.
lowbloodsugar: Google 10.9B gallons 2025, up 34%, Amazon 2.5B.
gorjusborg: thank you.
daedrdev: US uses 320B gallons a day, Google 0.01% of total US water, 0.6% of water for almonds.
phyzome: compare to planned datacenters.
daedrdev: US data centers going to 3x this year, still not a lot.
kragen: context: Google 10.9B gallons/year = 41.3M cubic meters/year = 1.3 m3/s, compare to Rio de la Plata 22000 m3/s => 0.006%; Lake Superior 1.207e13 m3 => 290k years to drain; Sacramento River 797 m3/s => Google 0.16%; agriculture in CA uses 34.1M acre-feet = 42.1B cubic meters/year = 1330 m3/s => 1000x Google; alfalfa 18% => 240 m3/s.
Need fresh water for crops, can cool DCs with salt water, but need 100x compute to be significant.
phyzome: AI apologist latch onto water thing ignore electricity, air pollution, noise.
daedrdev: electricity, air pollution, noise from natural gas turbines; can install solar/batteries.
zahlman: political/economic opportunity shift subsidies from corn/ethanol to solar.
phyzome: not talking hypothetical; current DCs use single-stage gas turbines; HVAC noise.
scarmig: stop spreading deception about water usage if bothers you; trolling to bring it up.
phyzome: I don't bring up water thing.
sandinmyjoints: human resources zero-sum game, agents silo not help cross-functional collaboration.
BadBadJellyBean: tired of directing agents like herding toddlers, constant nudging, token budget, losing brain power.
AIiscoming: AI builds 95% for me then I play around 5%.
BadBadJellyBean: dislikes telling people what to do, AI exclusively like telling someone super smart or dumb, need to proofread for style.
gonzalohm: conflict of objectives: AI companies need you to use more tokens, they optimize to get close to useless point.
zamadatix: conflict fine, not single provider.
sifar: every provider incentive for users to consume more tokens.
swader999: Welcome to management.
BadBadJellyBean: humans don't switch between idiot and genius fast as LLMs, thinking about future, not cut out for manager.
chasd00: using agents to make changes, got it mostly right, validation green, would not have happened without agents.
BadBadJellyBean: don't deny value, but if works well great, if not exhausting, if produces bad code overwhelming.
Refreeze5224: great but does not justify social and environmental costs of AI (point of article).
bunderbunder: would not have happened back in day, sensible leadership would talk delays adjust schedule.
archagon: āLook at how much more busywork I can do for the same salary!ā
cyberax: review agentic code, commit myself, write some features manually, slow down dev on purpose helps.
RomanKornev: herd a group of toddlers = steering.
zahlman: point at which you can take over, they saved time while not causing agony?
kragen: post not well-thought-out, pro/anti both simplistic, water-wasting red herring, environmental doom-mongering, labeling recursive self-improvement as mystical.
danvayn: not criticizing, just pointing out not normal HN post, journalistic blog, emotive, personal meaning lost if well thought out.
kragen: hate from dysfunctional responses, spreading hate online not socially positive.
donbox: front page some value.
kragen: front page engaging not valuable.
jakzurr: Amen.
voidhorse: OP raises broad social/ethical concerns about AI, provides links, vents frustration, personal anecdote, not unreasonable.
queenkjuul: Yes, everyone who disagrees with you is unhealthy.
ghostbrainalpha: what do you mean by firehose quality?
mikestorrent: quick to produce, emerges in vast quantities, consumes attention.
kragen: example of parsing engine, learning OCaml, etc.
simonw: love comment capturing experience of engaging with LLMs to challenge and expand knowledge.
kragen: guess you have well-thought-out blog post.
simonw: formulating idea, should write up.
kragen: hope you do.
jakzurr: Yes! Please do.
nlawalker: assume they don't have time to do that, lament exercising capabilities no longer on critical path to paid work.
kragen: suspect you still need them.
zero_shift: prototyping network proxies vibe coded, mixed experience, accelerated learning via writing rigorous plans, learned HTTP2, MITM proxy, SSL pitfalls, via Claude rendering diagrams; with Fable shipped Envoy Rust extension one-shot; grasp on details tenuous, agent can't progress, stuck for ideas, hitting prompts like slot machines, forget how to solve problems myself; becoming supine and dependent on something known to become more expensive, AI vendors will seek to exploit; happened quickly, do not really consent; forced by employer.
kragen: try typing in all code myself; Claude Pro only 1000 lines/day in 2-3 five-hour sessions; can type 1000 lines in couple hours if no problem solving; maybe look at Claude's code, switch screens, type until don't know what to type, repeat.
zero_shift: interesting idea like crib notes, forces pay attention, exercises mechanical memory.
jakzurr: Agreed, typing in takes less time than almost anything done during coding/dev.
keeda: best part is they are right there to answer questions, challenge or validate experiment with ideas; can go down rabbit holes; unlike any other medium other than live human expert you can hone in only on areas that matter; or ask to do something (digital realm) see what happens.
puchatek: Water-wasting not red herring if live in area where droughts are issue; not universal.
_kulang: need to live there, presence of data centre for AI which actually wastes water; numbers absurd; they don't take into account circular reuse of water for cooling, just look at flow rates.
simonw: Like golf courses in desert; Palm Springs ~100 golf courses.
kragen: while talking about Americans using anything but metric system, how many golf courses worth of water is Google's 10.9B gallons?
zahlman: ChatGPT says average 18-hole facility uses ~152.5 acre-feet/year = ~49.7M gallons/yr => about 220 golf courses.
But still drop in bucket vs CA almond industry.
simonw: satirically suggested hyperscalers buy golf courses turn to public parks.
runarberg: smokers in 1950s saw health benefits of tobacco, increased energy, no lung deterioration feeling.
Likewise, you may not feel like AI is de-skilling your brain but it probably is; we don't have research yet AI addictive causes brain atrophy, but evidence pointing that way same as tobacco evidence in 1950s; hope not another 2 decades to prove obvious.
kragen: feel strong pull toward low-effort path of letting AI do everything; code it wrote beautiful but makes terrible mistakes constantly; when reduced to pleading with it to check mistakes basically no hope; would put me in Homer Simpson position of declining to get up and walk because motorized wheelchair at hand.
PorciiVorbesc: It is, but why is that a problem? Moving from physical labor on farm/factory/construction to sedentary work/lifestyle would atrophy muscles so we pursue physical activities we enjoy like gym/sports instead of hanging onto dangerous strenuous labor just to stay in shape.
Similarly, you will use AI to churn through repetitive mundane tasks AI can do, then work your brain on activities you actually enjoy.
Now if you choose to not exercise your brain and dumb yourself down same way sedentary people don't exercise bodies and get obese, then that's on you, but we can't force world to technologically regress just because some people are lazy and don't exercise, same way we haven't forced construction workers to use pickaxes instead of machinery to tear up road just to stay in shape and not get too fat from sitting in air conditioned excavators.
zero_shift: > It is, but why is that a problem? I used to have several good answers, but now I use Claude for everything, I can't remember any.
johnnyanmac: >then work your brain on the activities you actually enjoy.
-
As seen in this coountry, most people did not in fact "pursue physical activities they enjoyed". They got much less healthy and then white collar machines started to churn their brains too. I do not see this ending any differently.
-
As many have noted (and has been known for decades pre-AI), getting more work done in a modern workplace does not reward you with more of your time back. It gives you more work and higher velocity plans. It may even end up compensating you less. There's only more benefit to this if you are your own boss, and so far a lot of those kinds of people are simply selling grifts, not truly disrupting the industry with newfound efficiencies.
but we can't force the world to technologically regress just because some people are lazy and don't exercise
I didn't want to go too off topic, but since this is such a sticking point I need to make a big note about the US: most of the world is more sedentary than 100 years ago. But the obesity epidemic is mostly US-cerntric.
Which shows that, while we can't force a diet, we can definitely regulate incentives. But the US chose not to. We had all kinds of initatives to try and regulate what we put in our food and what to subsidize. We instead said corporations can do whatever we want and subsidized chemicals that are heavily taxes or banned in many other western countries.
I know that's a scary word for this community, but this analog of regulations can work here in AI as well. We don't need to ban AI (but I'm not opposed to it in primary schools), but we can shape policy to prevent "corporations can do whatever they want" yet again.
PorciiVorbesc: >we can definitely regulate incentives.
We can't regulate incentives. Unless we can somehow go back to how the world was in the 1960s, the incentive today in a global competitive world economy, is to make more money than your economic rivals, otherwise if you only regulate yourself but if they don't, then they will economically dominate you in the long run and turn your government into their puppet, your country into their playground and your people in their sweatshop workers. It's economic neo-colonilism 2.0 electric boogaloo.
Regulations only work if all players in the world adhere to them, but history has proven they don't. They expect everyone else to adhere to the rules and slow down so that they can get ahead of you.
> But the US chose not to
Smart. Then US can be the survivor in this global race. As someone in the EU I am aware of more top EU-born tech companies relocating their offices/R&D to the US than vice versa.
> but we can shape policy to prevent "corporations can do whatever they want" yet again.
I fully agree. But again, this will work if all major nations regulate AI equally, otherwise, AI corporations will move where they have more freedom.
johnnyanmac: >We can't regulate incentives.
We call tax subsidies "tax incentives" for a reason. You subsidize high-fructose corn syrup, you incentivize society to include cheap oils in their food.
the incentive today in a global competitive world economy, is to make more money than your economic rivals
On a domestic level this is irrelevant. It doesn't matter how cheaply China makes corn syrup if there's still a cost of importing it into your country. And if it's still a problem, you add tariffs for more (dis)incentive. That was the original way they were supposed to be implemented, not as this massive general tax on doing any business with a country.
And if it's still a problem despite that, direct regulations and bans from usage in market consumption products is a last option. We can , do, and have done all of these for hundreds of ingredients.
Then US can be the survivor in this global race
I'm talking specifically about food here, you know? You were so focused on the obesity epidemic that