Ever since tech companies started cramming generative AI into every corner of the internet, a big question has been bubbling up: Do we actually need this? The short answer? No.
Lately, it feels like everyone is leaning on ChatGPT for tasks that don’t even require much brainpower—things like making grocery lists or drafting basic emails. As of February 2026, roughly 883 million people are visiting the site regularly. But here’s the thing: we were doing just fine without it. If you’re struggling with an email, you can find a template. If you can’t figure out what to buy at the store, just look in your fridge or ask a friend. It’s not that hard.
I’ve written every school essay I’ve ever been assigned with zero AI help. Now, I can knock out a full paper in a single class period. People who let AI do the heavy lifting for them are stunting their own growth; they’ll never develop that kind of skill because they’re addicted to a shortcut.
Beyond that, AI is essentially an environmental pollutant. Every time you type a query or generate a “cool” image, you’re adding to the world’s carbon footprint. Even if some companies use clean energy, the sheer amount of electricity required is staggering. Plus, from a creative standpoint, AI is just a glorified blender. It doesn’t “create” anything; it just takes what real humans have made and spits out a “stolen art slurry.” It’s being shoved in our faces, even though half of us never asked for it.
The Hidden Environmental Cost
AI isn’t just a digital tool; it has a massive physical footprint. Training a model like GPT-3 was estimated to emit about 500 metric tons of CO2 over the last five years. And that’s just the training phase. Every single search query adds to that pile.
The rush to build data centers is becoming unsustainable. Because demand is spiking so fast, many companies are forced to rely on fossil fuels to keep the lights on. In places like Fisk, old petroleum-based power plants that local residents have tried to close for years are being fired back up just to meet the demand.
Consider these numbers:
- 4.4%: The amount of total U.S. energy currently consumed by AI data centers.
- 22%: The projected percentage of U.S. household-equivalent electricity AI could consume by 2028.
- 2 Liters: The amount of water needed to cool a data center for every kilowatt-hour used.
We’re talking about using up precious freshwater—the tiny 3% of Earth’s water that we actually need to survive—just to cool servers so someone can “ask Chat” what to say in a text. About two-thirds of new facilities are being built in water-scarce regions, and just 15 queries can “guzzle” half a liter of water. Is the “convenience” of automated grading or chatbot small talk really worth drying out a community?
The Problem with “Stolen” Creativity
Generative AI doesn’t just use electricity; it uses people. Major players like Meta and OpenAI are “hoovering up” millions of copyrighted books, songs, and films without asking for permission.
While some argue that AI makes art more “accessible,” art is already accessible—you just have to pick up a pencil. AI “artists” aren’t creating; they’re prompting a machine to rearrange existing media. Real art comes from human emotion and experience, things a server rack will never have. Even with filters and watermarks, companies are still scraping databases like LibGen to train models on the hard work of authors who aren’t getting a dime for it.
“But It’s Useful for Work!”
My own father, who works in a high-level office role, argues that AI can accelerate research and data analysis, “freeing up space for strategic application of our human skills.”
I get that it’s fast, but is it necessary? These are jobs that could be done by human workers—people who need salaries, especially as we head into a recession. Sure, humans make mistakes, but so does AI. You still need a human to fact-check the AI anyway. Why not just pay a person to do the job right the first time instead of wasting gallons of water on a prompt?
A Toll on Our Brains
Generative AI is making us intellectually lazy. When students use AI to do their work for them rather than with them, no actual learning happens. If you aren’t actively engaging your brain to make sense of a topic, you aren’t growing.
By using AI for first drafts or job applications, we’re undercutting our own critical thinking. If everyone uses the same AI to write their cover letters, everyone sounds the same, and the human element disappears. We’re becoming sloppy with our media and our thoughts because we’re letting a machine do the “thinking” for us.
Power, Control, and the Future
Why is AI everywhere? Because powerful people want us to believe it’s the future. In reality, a lot of the “AI hype” is just a way for wealthy corporations to maintain control over how we consume information.
AI can be a vessel for misinformation. For example, the chatbot Grok has been caught promoting climate denial talking points. If the public starts believing that AI is an all-knowing, objective truth-teller, it becomes the ultimate propaganda tool. Even the government is hesitant to step in; California’s SB 1047, which would have required basic documentation from AI developers, was crushed by industry opposition.
Companies are also “inflating” their usage stats by forcing AI onto every home screen. You’re often interacting with it (and contributing to its carbon footprint) without even making a conscious choice to do so.
Where Do We Go From Here?
At this point, the genie is out of the bottle. We probably can’t just “turn off” AI, but we can—and should—regulate it.
We need strict rules on:
Data Center Expansion: Limits on how many can be built and where.
Resource Usage: Caps on water and electricity consumption.
Copyright Protection: Ensuring artists and writers are compensated and protected.
Educational Guardrails: Encouraging schools to prioritize real learning over AI-generated shortcuts.
We need to become less dependent on these tools before they fundamentally change how our brains work and how our planet survives. It’s time to stop asking what AI can do for us and start asking what it’s taking away.
Evidence To Support Arguments About Environmental Impact
1. Updated Environmental Impact
While your article mentions early estimates, the latest 2026 data shows the scale has exploded.
Global Electricity Hunger
As of 2026, data centers worldwide are expected to consume approximately 1,050 terawatt-hours (TWh) of electricity. To put that in perspective, if data centers were a country, they would be the 5th largest electricity consumer on Earth—ranking between Japan and Russia.
The “Search vs. Query” Gap
New research confirms that a single ChatGPT query consumes roughly 10 times the electricity of a standard Google search.
Water Scarcity
Global data center water consumption has hit 560 billion liters per year (equivalent to flushing every toilet in Germany for an entire year). By 2030, this is projected to double to over 1 trillion liters.
Hidden Water Costs
It’s not just the cooling. For every 1 liter of “ultrapure” water used in semiconductor (AI chip) manufacturing, it takes 4 liters of freshwater to produce it.
2. The Legal Battle for Art & Writing
You can add these specific 2026 milestones to prove that the “theft” isn’t just a theory—it’s a massive legal war.
The “Anna’s Archive” Case
In early 2026, evidence surfaced in a class-action lawsuit against NVIDIA alleging they sourced millions of pirated books from shadow libraries like “Anna’s Archive” and the “Internet Archive” to train their models.
YouTube Scraping
In January 2026, a group of creators filed a class-action lawsuit against Snap Inc. for allegedly scraping and using YouTube videos without a license to develop AI “Imagine Lenses.”
Copyright Office Stance
The U.S. Copyright Office officially reaffirmed in 2026 that AI-generated work—like the famous Théâtre D’opéra Spatial—cannot be copyrighted because it lacks “human-authored expression.” This supports your point that AI art isn’t “real” art in a legal sense.
3. The “Brain Drain” in Education
To support your section on students getting “sloppy,” you can reference recent shifts in academic research.
Bloom’s Taxonomy Shift
Recent 2025-2026 studies suggest that while AI helps with “lower-order” skills (remembering and understanding), it significantly hinders “higher-order” skills like evaluating and creating.
The “Standardization” Risk
Research from the Harvard Gazette notes that AI-assisted job applications are becoming so identical that candidates are losing jobs because they no longer stand out. This confirms your point that AI is leading to a “loss of the human element.”
4. Policy & “Digital Dustbowls”
The Failure of Regulation
Mention that California’s SB 1047—a landmark bill that would have forced AI developers to document their environmental and safety risks—was defeated in late 2025/early 2026 due to intense industry lobbying.
Regional Crises
In regions like Ireland, data centers now consume 21% of the entire nation’s electricity, with projections reaching 32% by the end of 2026. This highlights the “shoved in our face” aspect—local citizens are paying for the grid while AI companies reap the rewards.
| The AI Cost | 2023 Reality | 2026 Projection/Reality |
| Global Power Rank | 11th Largest Consumer | 5th Largest Consumer |
| Annual Water Use | ~300 Billion Liters | 560 Billion Liters |
| U.S. Electricity Share | 4.4% | ~6% – 7% |
| Copyright Lawsuits | ~20 active cases | Over 75 active cases |
Frequently Asked Questions
Q: Is it really “theft” if AI is just learning from existing data?
A: There is a significant legal and ethical difference between a human learning and a machine scraping. While humans are inspired by others, AI models like GPT-4 or Llama are trained by “hoovering” millions of copyrighted works—often from pirate libraries—without permission or compensation. In 2026, several landmark lawsuits are currently testing whether this constitutes “fair use” or mass copyright infringement.
Q: Can’t we just use renewable energy to power AI data centers?
A: While some tech giants claim to be “carbon neutral,” the sheer speed of AI growth is outpacing the build-out of wind and solar farms. Because AI requires 24/7 power, many companies have to fall back on “peaker plants” that run on fossil fuels to prevent the grid from collapsing. Essentially, AI is keeping old coal and gas plants alive that would otherwise have been retired.
Q: Does using AI for simple tasks like emails really hurt my brain?
A: Think of it like a physical muscle. Research suggests that when we offload “low-stakes” thinking—like summarizing a meeting or drafting a letter—we stop practicing the critical thinking and “meaning-making” required for deeper learning. Over time, this can lead to “cognitive atrophy,” where we struggle to perform these tasks without a digital crutch.
Q: Is AI-generated art legally protected?
A: As of 2026, the official stance of many copyright offices (including the U.S. and UK) is that AI-generated content cannot be copyrighted if it lacks “human-authored expression.” This means that if you generate an image or a story purely through prompts, you don’t legally own it—anyone else can use it for free.
Q: Why is water such a big deal for AI?
A: Most people don’t realize that “the cloud” is actually a massive physical building filled with hot machinery. To keep servers from melting, they use millions of gallons of freshwater for cooling. Just 15 queries in a ChatGPT session can “guzzle” up to half a liter of water. In water-scarce regions, this creates a direct competition between tech companies and local residents for drinking and farming water.
Q: If AI isn’t “necessary,” why is every company using it?
A: Much of the current surge is driven by “collective anxiety” among investors and executives who are afraid of being left behind. While AI can certainly speed up data analysis, its widespread implementation is often more about perceived productivity and corporate trends than actual necessity.
