As you know, the Anti-Dave has another life on the web at two financially oriented websites — UrbanSurvival.com and Peoplenomics.com. In that role I hear all kinds of feedback from regular people. This week a reader sent me a heartfelt screed that boiled down to a single, breathless claim: query AI and you ruin the planet.

“Ask a question and you use a gallon of fresh water,” she wrote from somewhere between a flourish and a fainting spell. I’ve heard the tune before — the Luddite aria of techno-guilt — and it’s worth walking through because the melody conceals a lot of bad economics and worse boutade.

Real AI Economics

Let’s take it from the top — and yes, a grounding in behavioral economics helps here. (If you don’t have one, don’t worry; I do.) Systems science teaches us to stop staring at single items like they’re the whole universe. Instead we draw concentric circles outward from the thing in question — think of life as ripples in a pond, not a collection of tidy, isolated pebbles. If I slug you, the first ripple is a bruise; the second might be a phone call; the third could be the cops. That’s the point: actions propagate. So when someone screams that using AI “kills a daisy,” they’re doing precisely what economists warn against — treating a high-order system as if it were a simple, one-step transaction.  Um…no.

Start with the obvious: what is the counterfactual? If you don’t ask the AI, what do you do instead? Type a query into a search engine, juggle four tabs, call a friend, flip through three books, or hire a consultant to come to your porch with a slide deck and a small, polite fee. All of those activities have real, measurable resource costs — time, fuel, paper, electricity, travel, and yes, water used indirectly in production and shipping somewhere along the chain. The anti-AI alarmists love to count the wattage behind a single cloud compute call while ignoring the entire supply chain of the human alternatives.

That’s not analysis; it’s cherry-picking with a moral blush.

But let’s play generous and accept the premise that a single AI query consumes energy. Good. So does leaving a lawn mower idling, running a truck to town for groceries, manufacturing another plastic widget, or printing another glossy brochure.

The right question — the one every decent systems analyst asks — is this: per unit of decision quality or useful output, which option uses fewer resources? If AI gets you the answer in two minutes instead of seven hours of human dithering, even a modest energy footprint looks efficient. Efficiency matters. Economists call this “opportunity cost.” If the AI stops you from making a bad, expensive, planet-hungry decision, you’ve just saved far more than the cost of the compute.

Decision Ripples Matter

Now push one circle out. Mistakes are expensive. A wrong purchase order multiplies upstream waste: shipping returns, landfill, extra fuel, worker hours burned on reversing a bad transaction. Bad medical triage decisions cost lives and dozens of machines running overtime. The “kill a daisy” crowd acts as if error rates are unaffected by the toolchain chosen.

Reality: better information reduces errors. Tools that raise the signal-to-noise ratio — and AI is a tool like any other — lower the chance of costly screwups. If AI shortens a trial-and-error loop, it reduces aggregate resource consumption.

Behavioral economics reminds us of another human quirk: people overweight visceral imagery. A photo of a wilted flower tugs at the heartstrings; a spreadsheet showing net resource use does not. That’s why the daisy metaphor spreads — it’s simple, emotive, and bad at arithmetic. We need to swap pity for math. If you really care about the planet, love the daisies, and sleep with a reusable water bottle under your pillow, you should ask: which workflows produce the least total waste over time? There’s the rub. Short, accurate, computationally cheap decisions trended by good models will usually beat long, error-prone human chains that culminate in expensive reversals.

The High Cost of Mistakes

A practical example from my newsletters: I’ll use an AI to triage market signals, to filter spammy pitch decks, or to sketch a first pass of an article. That saves hours I would otherwise spend chasing red herrings. Those hours are often spent commuting to a coffee shop, printing pages, and buying snacks while I grind through bad leads. Multiply that across thousands of users (or decades of life) and you’re looking at a lot of human activity that generates far more footprint than the back-end servers doing the heavy lifting. If the AI stops a fool from ordering 10,000 useless widgets from Shenzhen, it just saved prodigious carbon, manufacturing, and yes — a lot of daisies.

None of this is a free pass. The AI industry should measure and own its footprint. (Then again, so should government.)

Data centers can and must get more efficient. Renewable power purchases and careful model design are obvious levers. But the moral panic that equates asking a question with ecological suicide is a rhetorical stunt, not a policy. It absolves the real levers of waste: throwaway culture, inefficient logistics, endless shipping, and the marketing engines that manufacture demand for stupid stuff. If you want to save flowers, start by stopping stupid purchases, not by banning curiosity.

Finally, there’s a deeper philosophical angle. Technology has always asked the same question: will it amplify human judgment or will it amplify human error? Fire can cook a meal or burn a house down. The net effect depends on how wisely we use the tool. AI, properly governed and applied as an assistant — not a panacea, not a replacement for judgment — tends to increase the former and reduce the latter. That’s not abstract optimism; it’s applied systems thinking.

Bet Me?

So here’s a small, practical wager: next time someone tells you that using an AI kills a daisy, ask them to account for the full ripple. Ask them what would happen instead, and how many human hours, shipping miles, and reversed purchases would follow. Ask for the baseline. When you see the arithmetic, the florid rhetoric usually wilts faster than a daisy on a Texas August afternoon.

In the meantime, keep asking questions. Curiosity costs kilowatt-hours, sure — but it also buys better choices. If we’re honest about the tradeoffs and measure the ripples instead of the splash, we’ll find the daisies have a better chance of surviving the accounting.

Now, about those Mind Amplifiers in more depth, shall we?

It occurred to the Anti-Dave long ago — because of all that systems grounding — that AI wasn’t some foreign invader of the intellect but the next logical step in how humans extend their own cognition. Before the year’s out, the book Mind Amplifiers will be in print, but there’s already a website up for it. The concept’s simple but profound: these things we call “mind amplifiers” are prosthetics of human cognition. Every generation builds better ones, just as we built better cars from the Model T forward.

Mind Amplifiers Explained

Such tools come in flavors. There are internal amplifiers — your own ways of seeing and processing the world. There are hybrids — like coffee, which begins outside the body but becomes an inner catalyst for focus. And there are pure externals — keyboards, books, calculators, AIs — all mirrors reflecting thought back into the mind. The trick is learning to use each class consciously instead of passively drifting through them.

Riffing off Julia Childs, “First you make a roux…” most human’s don’t start with their objectives and work back to tool selection.  Nope.  We fall in love with a tool first.  Which is how hammers went looking for nails and suddenly every problem looked like a nail…

Internal amplifiers shape perception itself. Some people operate as “tunneling silos,” burrowing deep in one mental channel and defending it to the death.

Others are “domain walkers,” able to move between ideas, frameworks, and disciplines without losing coherence. The latter make better sense-makers because they see relationships instead of walls. How you use AI — and how you interpret something like the “kill a daisy” meme — depends on which kind of thinker you are. Silo minds fear tools that connect domains; domain walkers welcome them.

Hybrids, like caffeine or even music, change the brain’s chemistry to open a window for sharper thinking. They’re ancient forms of amplification — monks had incense and chants; engineers have espresso and playlists. The line between natural and artificial was always arbitrary. What matters is the quality of the amplification, not its origin.

And the externals — your computer, your smartphone, your AI assistant — are just the latest mirrors held up to consciousness. Every time you type, read the reflection, and adjust your thought, you’re in a feedback loop. That’s what intelligence is: reflection plus iteration. When someone rails against AI as if it’s a separate species, they’re missing the continuity. The tools are us — extended, scaled, and looped back through silicon.

So we don’t argue with the folks wringing hands over “super AI risk.” They’re fighting a cartoon version of reality. The real risk has always been human: cars, guns, drugs, governments, and yes, doctors when they get it wrong. If you want to talk about planetary hazards, those are the big leagues. Daisy-killers they are not.

In the end, daisies don’t die because someone asked a question. They die when curiosity is replaced by fear, when people stop thinking in systems and start reacting in slogans. And that’s why Mind Amplifiers matter — not as machines, but as reminders that the mind itself is still the most powerful renewable resource we’ve got.

~ the Anti-Dave