• Top
TLCR: Why are those, most openly opposed to AI, are such avid users of it, and the direct investors into it? That is what will be answered and explained herein!
AI and/or AI Data Centers would not even exit (not now; not in 2040), if not for the presence of (now) 7.5 billion smartphones. It's analogous to every smartphone user being one piece of a very large puzzle, who then condemns that puzzle: That defines hypocrisy!
But very few consumer drones know anything factual about AI or their Data Centers; read and comprehend this one site and you'll then know a small percentage of what I did about this subject 5 years ago.
These are sheeple herds, living contrarianism and then blindly following others online (somewhere), only to join any beat-down of "other" that will give them that fleeting moment of self-worth. They have no genuine stance on anything of value. Our world is haunted by herd mentality.
In reality, AI Data Centers are the culmination of investments made by consumers who used smartphones over 2 decades (2006 to 2026). Born in 58, I did not even have one until Oct. of 2016 and had 4 until Oct. of 2025 when I reversed back to Linux-based everything, including a Go-Bag Laptop that I take everywhere. So I only use text-based AI searches, and have never been a major contributor to any of it.
Everyone with a smartphone or a PC or laptop or tablet or smart TV or smart appliances or talk boxes (Alexa, Siri, Gemini, etc.); these all use Artificial Intelligence (AI), as well as more than one Data Center at a time, and often, and everyday. If your smartphone is from Android or Apple, you add yet again (and heavily) to the financing of the businesses that are building AI Data Centers. Add to all of that, any Tech Billionaire Social Media app, and you're a power-AI user and a super-contributor to AI Data Centers! FYI!
Though tee AI boom would not have been possible without billions of smartphones, their role is distinct from the hardware that computes AI. Data Center GPUs provide the necessary compute power to train models; smartphones provide the essential data fuel and user interface that made modern AI viable. Phones serve as the primary interface for AI (directly & indirectly): Voice assistants, real-time translation, and computational photography, demonstrated AI's utility to the mass market, driving the investment fueling the boom.
In 2026, the 4 largest Tech companies; Amazon, Alphabet, Microsoft, Meta, committed a combined $700B to AI infrastructure, representing the largest single-year corporate investment cycle in history, with Alphabet and Amazon planning $200 billion each. They get all their monies from the customers of their products and services, so if you patronize any of these corporate entities, congrats, you're an indirect investor in these AI Data Centers!
It's wildly hypocritical to be on a smartphone, plus on any social media app, then condemning AI in any way. Well, unless you're just a hater, seeking other haters to gang-up on today's enemy, such as the current hate all AI fad. None of this Tech would be possible without you, times billions of you. I'd wager large that if your smartphone started failing badly, you would raise hell. So this is about making enemies by being conspiratorial, which has always happened with all new tech experiences throughout all time.
In order to be reading this (or anything online) you're using a conventional or hybrid Data Center right now. If you're on a smartphone you are connected to many of them, and nonstop and all the time. This connection in perpetuity, times 8 billion phones, is added to all other always-connected devices; this drove the first great wave of Data Center construction. By 2024 there were 12,000 operational Data Centers globally; all were built unnoticed by the simpleton masses of haters and fakers.
But, since everyone was responsible, and there was yet to be a common enemy circulating and parroted by the enemy-seeking hatemongers, out hating all over social media, no one objected. It's easy to make an enemy out of the Great AI Beast, and yet, that same everyone is why it exists at all. What AI ultimately can accomplish is so very far beyond the comprehension of the sheeple masses, they cannot do anything else to follow others to whatever discriminatory beat-down is occuring in real-time in their presence.
Consequently, haters don't really care about anything factual; they exist but only to spew the venom of hate always! And the hate core comes from their pride, which itself is overcompensation for their own stupidity. Like, they will use AI searches such as Google, or any other Big Tech search entity, which all use AI within AI & hybrid & conventional Data Centers, to find out if any of this it true! I know, wildly hypocritical! Embarrassing even, to those with confirmed IQ's North of 110!
The Car Cult had done more damage to Earth by 1970 than all Data Centers on Earth forever will do, over their entire lifetime's of operation. It's not really comparable in scale; the Car Cult does more damage to Earth every 6 months than all Data Centers on Earth forever, will do in forever. Data Centers and AI will not end us as a species, but cars absolutely will.
The entirety of all Data Center Infrastructure that will exist on Earth from 1990 to 2090, will still not represent a fraction of all human infrastructure that's not designated as human shelter. Whereas, 98% of all human-built infrastructure (outside of human habitats) that scars Earth, is for cars, though AI Data Centers are more expensive per their size or square footage of infrastructural space.
AI Data Centers globally, built between 2025 and 2050, will still not cost as much in spent nonrenewable natural resources, as does just the snow removal and repair of existing roads (circa 2025) during the same timeframe. AI Data Centers can be ran from nonrenewable resources, and their upkeep is minuscule compared to global roads kept in perpetuity.
So, the only people who care about Earth, do not own or use cars! Condemning anything else is just a distraction from this rampant devastation. Those who are truly against any such Tech advancements, are those who DO NOT have any such tech devices or services; not even a smart TV! To otherwise oppose anything in Tech world, is just online victim play. (End of TLCR: I know that was hard for you ;–)... Researchers may continue. See: AIDC.txt.
As an advocate for Small and Low Tech, and especially living a Car Free lifestyle, this lifestyle proves my/our stance as Lifestyle Environmentalists. That, as opposed to the talk-only people, who need a fake online presence so they can pretend to be the person they are not IRL! Cars represent most of their owners carbon footprint. But the rapid development (or race really) to build AI Centers are certainly creating ecological issues of their own.
Artificial Intelligence (AI) is financed primarily by Big Tech giants, and they get all their money from their customers and account holders and other click-enablers who establish ad revenue dollars. The Big 4 AI hyperscalers are Alphabet, Amazon, Meta Platforms and Microsoft. Getting all Development moneys from customers and/or account holders, it's bizarre to hear/read their hate of what they finance, buy, and make possible.
Microsoft with Cloud services, Windows OS, and hardware such as Surface and Xbox. Meta Platforms account holders, all create the ad revenue that Big Tech uses to create AI and modern Data Centers. And Alphabet has Cloud services, Google search, and YouTube as their main moneymakers. But with the sheeple numbers making smartphones for dump people, AI and Data Centers are that inevitable technological advancement, just like those that came before.
The Big-4 plus Apple, form the core group of U.S. Big Tech driving AI innovation, with Alphabet, Microsoft, and Meta often cited as the Big-3 in terms of AI infrastructure spending, model development (e.g., Gemini, Copilot, Open AI, Llama, etc.), and enterprise cloud market presence.
Therefore, those who buy from, or have accounts with, or use, any of these Big Tech Big Biz products and services, they are the direct financial supporters of all major AI investments, with particular focus on modern Data Centers. The greatest uses by volume of Data Centers are enterprise applications, which accounted for approximately 55% of total power demand, driven by file storage, transaction processing, and business digitization.
The next largest categories include Cloud Services, which facilitate scalable, on-demand applications and infrastructure for businesses. Then it's Content Delivery: Streaming video and online services that involve videos and images; these require massive distribution networks to keep content near end-users.
And last in this list of resource consumption lineup, is AI, the new Tech that gets all the condemnation from the online beat-down bullies. While growing rapidly and driving new construction, AI currently represents a small volume of total existing Data Center capacity compared to traditional enterprise workloads.
But AI has long been an inevitable step in computer advancement, so there is nothing particularly conspiratorial about it. Us elders remember all the warnings about computers themselves, and the devil's Internet: It was going to bring a literal Apocalypse to humanity and we would not survive much past 2020 at most!
But this is almost as old as time; anything introduced as new, will stumble those who are intellectually and/or developmentally disabled, and this is proven by how they confront anything new; with rejection and/or skepticism, and without any usable research capabilities. It's most likely that those who villianize AI and/or Data Centers, are themselves the main revenue sources that are building it all.
But extremism and spam messaging, will lead far more sheeple online, than logic or reason or data ever does! The current "No–AI" surge of down-voters bears witness to this fact. It is ironic because those who are "No–AI" are doing so on sites that are heavily dependent on AI as well as on Data Center natural resource usage.
One SUV over its 25 year lifespan will do more ecological damage than the largest Data Center will within the same timeframe: And there are multiple billions of them! If one person used AI for 12 hours everyday for 60 years, that will still not do the same damage their car does every 6 years. But here's the thing; they and everyone they know, has a car, so it's fully normalized in their lives.
Indeed! Most people don't actually care about the future of Earth; they care only about their own ego, as the overwhelming evidence teaches us learners. So being discriminatory gives them the illusion that they're better than whatever "Other" they choose to oppose, but since they suffer from a low IQ (all discrimination is empirical evidence of a low IQ), they overcompensate via the "I'm better then you" paradigm, which they live out in public life, and even openly when they're anonymous like when online as some fictional character they fake, all equipped with fake morals, ethics, and so on.
Many social media users criticize AI-generated content while unknowingly relying on AI-driven features daily. Below are the primary forms of everyday AI integration that fuel this paradox.
Everyday AI Forms Used Daily!
1. Smartphone Functions:
Modern smartphones utilize AI for spam call detection, photo organization (grouping by faces or scenes), automatic spelling and grammar correction in messages, email recipient suggestions, and traffic-aware routing in mapping apps. These features operate via machine learning algorithms embedded in the device or carrier systems.
Everyday AI features in Smartphones!
2. Social Media Feeds and Ads:
Platforms like Facebook, Instagram, TikTok, and X use AI to curate personalized content feeds, target advertisements, and recommend accounts to follow. These systems analyze user behavior, engagement patterns, and preferences to optimize content delivery.
3. Streaming Recommendations:
Services such as Netflix, Spotify, and YouTube deploy AI to suggest movies, shows, songs, and videos based on viewing history, completion rates, and similarities to other users’ preferences. These recommendation engines shape much of users’ media consumption.
4. Financial Transactions and Credit:
AI monitors credit card transactions in real time to detect fraud, analyzes spending patterns for loyalty programs, and automates credit approval decisions by scoring applications for risk. Loan assessments for cars and homes also increasingly rely on machine learning models before human review.
5. Travel and Transportation:
Usage-based car insurance tracks driving behavior via AI to adjust premiums. Toll systems use optical character recognition to bill vehicle owners automatically. Rideshare apps like Uber and Lyft apply AI for driver assignment and dynamic pricing. Airport security employs facial recognition to verify identities against IDs.
6. Health Monitoring:
Smartwatches and fitness trackers use AI to monitor sleep quality, detect irregular heart rhythms like atrial fibrillation, and provide fall detection alerts. These health insights are generated through continuous machine learning analysis of biometric data.
7. Online Shopping and Marketing:
Retailers apply AI to analyze purchase histories, predict product affinities, and personalize coupons or promotions. E-commerce platforms use recommendation engines similar to streaming services to suggest items based on browsing and buying behavior.
8. Content Creation Tools:
Even critics of AI often use grammar checkers, auto correct, photo enhancement filters, or voice-to-text features; all powered by AI. Some may also leverage AI indirectly when platforms auto-generate captions, translate posts, or suggest hashtags.
Unconscious use of AI content tools:
These pervasive AI applications illustrate the contradiction. While many denounce AI-generated posts or comments, they simultaneously depend on AI for communication, entertainment, finance, travel, health, and commerce.
Executive Summary!
Video streaming and image generation consume significantly more energy and data than text-based activities. While a single text query is relatively lightweight, high-bandwidth activities like HD video streaming and generative AI video creation dominate current data center usage. Streaming video alone accounts for 60–70% of all global internet traffic, whereas text-based interactions represent a tiny fraction of total load despite higher per-query energy intensity compared to simple searches.
Energy Consumption: Text vs. Image vs. Video!
The energy cost of digital activities varies drastically depending on the media type. Text is the most efficient, while video and generative media are the most intensive.
• Text-Based Activities: A standard Google search consumes approximately 0.0003 kWh (0.3 watt-hours) per query. In contrast, a generative AI text prompt (like ChatGPT) is more intensive, using roughly 0.003 kWh (2.9 watt-hours); about 10 times more energy than a traditional search. However, this is still minimal compared to media streaming. An average AI text response uses about 114 Joules, which is comparable to powering a microwave for a fraction of a second.
• Image Generation: Generating a single AI image is substantially more demanding than text. An average image generation request consumes approximately 2,282 Joules (roughly 0.003 kWh). While this is higher than a text query, it remains lower than sustained video streaming on an hourly basis.
• Video Streaming and Generation: Video is the most energy-intensive common activity. Streaming one hour of HD video (e.g., Netflix or YouTube) consumes between 0.036 kWh and 0.12 kWh, releasing roughly 36–55 grams of CO₂. To put this in perspective:
• One hour of Netflix is energetically equivalent to approximately 40 to 400 AI text queries, depending on the specific study and model complexity.
• Generative AI Video: Creating a mere 5–10 seconds of AI-generated video online can consume 0.05 kWh to 0.94 kWh, rivaling or exceeding the energy cost of watching an hour of streamed content.
Data Center Load Distribution!
While AI is growing rapidly, traditional workloads like streaming and social media currently dominate data center energy consumption.
Current Share As of 2024, AI-specific servers accounted for an estimated 53–76 TWh of electricity, representing roughly 14% to 20% of total global data center usage. The vast majority of the remaining load is driven by video streaming, cloud gaming, social media, and enterprise cloud services.
• Traffic Volume: The disparity in load is largely due to data volume. Video files are exponentially larger than text.
• Streaming Dominance: Video streaming constitutes 60–70% of all global internet traffic.
• Text Efficiency: A whole conversation's worth of text might total a few kilobytes, whereas a single second of video contains vastly more information to transmit and process.
The Impact of Streaming Sites!
Streaming platforms like Netflix, YouTube, and Zoom are the primary drivers of data center energy demand due to their continuous, high-bandwidth nature.
Comparative Footprint!
• Netflix/YouTube (1 hour HD): ~0.12 kWh → 42g CO₂. This is cited as one of the "dirtiest" single digital activities.
• Zoom (1 hour): ~0.05 kWh → 17g CO₂.
• AI Text (2 prompts): ~0.00024 kWh → 0.084g CO₂.
Research indicates that watching Netflix for one hour creates roughly 500 times more CO₂ than sending two text prompts to an AI chatbot. Even compared to image generation, an hour of streaming generally exceeds the energy cost of generating several individual images. The "rebound effect" further complicates this; while streaming efficiency has improved by ~20% annually since 2010, total viewing hours have risen enough to offset these gains.
Future Projections and Trends!
The gap between text and media usage is widening as AI capabilities expand into video and as global streaming habits grow.
Growth Rates!
• AI Growth: Electricity consumption from AI-accelerated servers is projected to grow at 30% per year. By 2030, AI could account for 40% of the generative AI market share and a significantly larger portion of data center load.
• Total Data Center Demand: Global data center electricity consumption is expected to roughly double to 945 TWh by 2030.
• The Shift to Video AI: While current AI usage is dominated by text inference, the emergence of text-to-video models represents a critical tipping point. Generating AI video is described as a "very heavy lift" that burns data center energy rapidly. If AI video generation becomes as ubiquitous as text chat, AI's share of data center usage could rival or surpass traditional streaming. However, for the foreseeable future, streaming video remains the single largest consumer of data center energy and bandwidth.
Is That Digital Art AI Generated?
To determine if digital art is AI-generated or assisted, examine the artwork for anatomical inconsistencies, texture anomalies, and contextual clues.
Visual and Technical Indicators!
• Anatomy and Details: Look for misshapen hands (extra/missing fingers, fused digits), inconsistent lighting (multiple conflicting shadows), and warped patterns where textures or clothing blend illogically into the background.
• Texture and Surface: AI art often appears unnaturally smooth, lacking the visible brushstrokes, canvas texture, or organic imperfections found in human-made work. Skin may look plastic-like and pore-less, while hair may appear too uniform or fused with objects.
• Composition and Logic: Check for impossible architecture (stairs leading nowhere, mismatched perspective) and generic poses. AI often lacks a cohesive narrative or intentional emotional depth, resulting in an "uncanny valley" effect.
Contextual and Provenance Checks!
• Artist Profile: Investigate the creator’s history. A sudden surge in high-quality posts without visible skill progression, or a lack of work-in-progress (WIP) videos, suggests AI use. Established human artists typically maintain a consistent portfolio and engaging community interaction.
• Metadata and Watermarks: Inspect image properties for creation software information. Some generators leave subtle marks, such as DALL-E’s colored squares or Google’s SynthID watermark.
• Reverse Image Search: Use tools like TinEye or Google Lens to trace the image’s origin. If the image appears in AI collections or has no credible provenance, it is likely synthetic. Verification Tools While no method is foolproof, you can use AI detection tools (e.g., Hive Moderation, Illuminarty, WasItAI) or art recognition apps (e.g., ArtScan) to analyze pixels or match against known human-made artworks for verification.
Study: https://search.brave.com/ask?q=list+the+most+to+the+least+energy+efficient+among+every+major+AI+Data+Center+brands
I'm an eye witness to how society hated computers in their early release years, then the Internet, and now AI. Yet, everyone who opposes AI, are the financial drivers of all of it, unless they are 100% analog, like I was for half my life! And I doubt that; they just represent those who love to hate and hate to Love! Disruptive Innovation has always just that; disruptive!
But AI just represents a fundamental shift from general purpose computing to specialized intelligence, driven by the transition from CPUs to GPUs, and Neural Processing Units (NPUs) designed for parallel matrix operations. Unlike previous hardware generations that merely accelerated existing software tasks, AI hardware enables neural networks to learn from data, compressing a decade of computing evolution into a few years through super-accelerated gains in raw compute and algorithmic efficiency.
Current trends indicate a move toward Personal AI (PAI), where intelligence shifts from centralized Data Centers to edge devices, allowing for local, context-aware processing that improves privacy and responsiveness. As AI agents begin to automate complex workflows in fields like medicine and scientific research, the technology is evolving from a tool for information retrieval into a collaborative partner that actively participates in discovery and decision-making processes.
As a Linux Geek, this long-time Small Tech, Open Source, FOSS World inhabitant, only uses AI deliberately via Brave Search. Keep in mind that if you use any of the Big Biz services mentioned herein, you are a heavy user of AI and Data Centers, as well as the accompanying natural resources that run them.
It is self-righteous and hypocritical to denounce AI and Data Centers, all while being a financier, via your investment in any and all Big Tech entities. It's bizarre that the average person will have a Windows OS PC and/or laptop, an Apple or Android phone, Cloud accounts, a Facebook account, be an avid Melon Husk via TwitX account holder and supporter, and use video and image heavy sites, and then, act (hypocritically) as innocent victims of AI and Data Centers. Embarrassing!
Conclusion: In traceable fact, such a blind consumer is as complicit in their investment in AI and Data Centers, as well as their great support of all the Tech Boy wealth-class, as is technically possible, and there is little else one could do to display a greater collaboration and devotion to it all, than what their lifestyles represent right now. It is all blatant hypocrisy!
But perhaps being trapped in (and led by) ones pride and ego and emotions, all self control has been lost. In provable fact, even if one used AI and Data Centers religiously and all day long, car use will still represent well over 95% of ones ecological footprint! The Car Cult is ravaging Earth. A half-million Data Centers running AI nonstop for a century won't do 1% of the ecological damage the Car Cult does yearly and has been for 8+ decades.
There Is No Such Thing As A Car Owning Environmentalist!
Here's my favorite youtube topics & personalities (I don't use YouTube very much):
https://www.youtube.com/@LaurieWired
https://www.youtube.com/@NotJustBikes
https://www.youtube.com/@MackExplains7