A.I. Is Eating the World
A field guide to a world mid-consumption — the decay, the machinery, the economics, the human wreckage, and the narrow path forward.
A.I. Is Eating the World
A famous tech philosopher once said, "Software is eating the world."
Software ate the world by turning physical things into digital ones. AI is doing something harder by automating the systems behind those things. Software replaced the filing cabinet, but AI is replacing the employees who kept the business running.
This essay pulls together the conversations already happening across the web about the chronological state of AI and what it's doing to everything it touches. It's meant to be a sketch map of the discourse.
Part I: The Digital Decay: The Memes-to-Slop Ratio
Organizations like the Internet Architecture Board believe that the internet is meant to serve the end users. Others, like tech company executives, believe that the internet is for porn. AI is burying it under a tsunami full of AI dogshit. This section maps the collapse of digital trust and content integrity.
1. An Avalanche of Slop: The Internet Is Drowning in Low-Effort Noise
The cost of producing content is free-falling to zero. The result isn't abundance, it's a cesspit. SEO farms, AI-generated scams, synthetic product reviews, auto-generated news articles: the web is filling up with text that technically says something but communicates nothing. We're past information overload. This is information contamination.
2. The Ghost in the Browser: The Dead Internet Theory, Realized
What was once a conspiracy theory is becoming measurable reality. Bot traffic now rivals human traffic. Comment sections talk to themselves. Social feeds are populated by AI personas engaging with AI content. The "dead internet" isn't a prediction anymore — it's an observation. The question is how much of the web is still alive.
3. The Ouroboros Effect: Can AI Survive on a Diet of Its Own Synthetic Data?
AI models are increasingly trained on AI-generated data. The research calls it "model collapse" — a slow degradation of quality as models feed on their own outputs. It's the intellectual equivalent of inbreeding. If the internet is the training ground and the internet is now mostly synthetic, what happens to the next generation of models?
4. The Death of Evidence: Navigating a Post-Truth World of Infinite Forgery
Deepfakes, synthetic voice, fabricated documents, AI-generated "photographs" of events that never happened. We're entering a world where any piece of media can be plausibly fake. The cost of forgery has dropped to zero. The cost of trust is about to skyrocket. Courts, journalism, and public discourse are all unprepared.
5. The Open Web is Dead. Long Live Walled Gardens
The public web has been rotting for a long time. Long before generative models arrived, SEO spammers had already wrecked the ecosystem. AI didn't start the fire, it was the final accelerant. The era of "surfing" is over. Kids aren't surfing the web, but they live entirely inside their favourite apps. A strange inversion has happened. The "walled gardens" we used to criticize are now our best defense. All that information siloed inside Discord servers, group chats, and closed communities? That's not a bug, it's a feature! In these invite-only spaces, people find something the open web lost, which are human curation, human trust networks, and human judgment about what's worth knowing. This points to something deeper about where the internet goes from here. The near future isn't better algorithms or smarter search engines, it's people. AI will eventually get better at curating the web very soon. But for now, the most reliable signal on the internet has always been the same. Someone you trust telling you something is worth your time.
Part II — The Engine Room: The Physical and Economic Cost
AI doesn't run on vibes. It runs on silicon, electricity, water, and money — staggering amounts of all four. This section exposes the physical infrastructure and resource politics behind the hype.
5. The Grid Breakers: Feeding AI's Multi-Gigawatt Hunger for Energy and Water
A single ChatGPT query consumes roughly ten times the electricity of a Google search. By 2026, AI's energy demand is projected to rival the entire power consumption of Japan. Data centers are draining local water supplies for cooling. The environmental cost of intelligence-on-demand is becoming impossible to ignore — and impossible to sustain at current trajectories.
6. Silicon Sovereignty: Compute as the New Oil and the Geopolitics of Chips
Whoever controls the GPUs controls the future. The semiconductor supply chain — concentrated in Taiwan, choked by export controls, dependent on a single Dutch lithography company — is now a first-order geopolitical concern. Nations are treating compute capacity like they once treated oil reserves. The chip war isn't a metaphor. It's industrial policy.
7. The New Robber Barons: How Wealth and Power Consolidate in the Latent Space
Training a frontier model costs hundreds of millions of dollars. Running one costs billions per year. That means the AI economy has a natural tendency toward monopoly — a handful of companies with the capital, data, and compute to compete, and everyone else renting access. We've seen this concentration before. We called them railroad barons, oil magnates, and tech giants. The new version moves faster.
8. Algorithmic Arteries: How AI Is Rewiring the Global Supply Chain
From demand forecasting to warehouse robotics to autonomous last-mile delivery, AI is reengineering every link in the supply chain. The gains in efficiency are real. So are the risks: increased fragility, single points of algorithmic failure, and the quiet displacement of millions of logistics workers who won't see it coming until the pink slip arrives.
Part III — The Future of Value: Work, Wealth, and Ownership
AI doesn't just automate tasks — it redefines what "valuable work" means. This section examines how labor, professional expertise, intellectual property, and financial markets are being reshaped.
9. The Developer's Paradox: From Manual Labor to Algorithmic Orchestration
The same tool that writes your boilerplate can write your replacement's first PR. AI is simultaneously the greatest productivity multiplier developers have ever had and the most credible threat to their livelihoods. The developers who thrive won't be the ones who code the fastest — they'll be the ones who learn to orchestrate, verify, and direct AI output. The job title stays the same. The job doesn't.
10. The End of the Middleman: Why AI Is Toppling Professional Gatekeepers
Lawyers, accountants, radiologists, financial advisors — professions built on licensed expertise and information asymmetry. AI doesn't need a license. It doesn't bill by the hour. When a model can draft a contract, read an X-ray, or file a tax return at 99% accuracy for 1% of the cost, what is the licensing regime actually protecting? Competence — or cartel?
11. The Prompt Is the Product: Who Owns the Output of a Billion Parameters?
Copyright law was written for humans creating from scratch. AI creates from everything it's ever ingested. When a model produces an image "in the style of" a living artist, or generates code that mirrors a copyrighted codebase, who owns the result? The person who prompted? The company that trained the model? The creators whose work was consumed without consent? The legal system has no good answers yet. The lawsuits are already here.
12. The Flash Economy: High-Frequency Markets in an AI-Driven World
AI-driven trading already dominates equity markets. Now it's moving into commodities, real estate, and credit. Algorithmic speed creates efficiency — and flash crashes. AI agents trading against AI agents in microsecond loops creates a financial system that humans can observe but no longer meaningfully control. The market is becoming a machine talking to itself.
Part IV — The Human Redefinition: Society, Agency, and Identity
Technology doesn't just change what we do — it changes what we are. This section confronts the deeply human consequences: how we learn, how we decide, how we connect, and who gets left behind.
13. The Socratic Crisis: Rebuilding Education for the Age of the Instant Answer
The essay is dead as an assessment tool. Students can generate A-grade work in seconds. But the deeper crisis isn't cheating — it's that the process of struggling to articulate an idea was the education. If the answer is always available, what happens to the capacity for original thought? Schools aren't just fighting AI. They're fighting the obsolescence of their entire pedagogical model.
14. The Autonomy Illusion: When "AI-Assisted" Becomes "AI-Decided"
It starts as a suggestion. Then a default. Then a decision you didn't realize you'd stopped making. Automation bias — the tendency to defer to machine recommendations — is well-documented and accelerating. From medical diagnoses to criminal sentencing to hiring decisions, "the AI recommended it" is becoming the new "I was just following orders." Human agency is being eroded one convenient default at a time.
15. The Synthetic Soul: Love, Loneliness, and Connection in the Age of AI Companions
Millions of people are already forming emotional bonds with AI chatbots. Some find genuine comfort. Others are retreating from human connection into relationships that are frictionless, always available, and entirely synthetic. When a machine can simulate empathy better than the average human interaction, what does that say about the machine — and what does it say about us?
16. The Global GPU Gap: Digital Colonialism and the New Divide
AI capability is concentrating in a handful of wealthy nations and corporations. The Global South risks becoming a consumer of AI systems it didn't build, can't audit, and doesn't control — systems trained on data that may not represent its people or priorities. The digital divide isn't just about internet access anymore. It's about who gets to shape the intelligence layer that will mediate everything.
17. The Regulation Race: Can Governments Leash a Machine That Moves at Light Speed?
The EU's AI Act. US executive orders. China's algorithmic governance rules. Every major power is scrambling to regulate AI, but the technology evolves faster than legislation. The result is a patchwork of inconsistent rules that simultaneously constrain responsible actors and fail to contain reckless ones. Regulation isn't optional — but getting it wrong might be worse than getting it late.
Part V — The Frontier: Where AI Goes Next
AI isn't slowing down. It's expanding into domains we haven't fully reckoned with. These are the next frontiers — the places where the "eating" is just beginning.
18. The Architect of Life: How AI Is Solving Biology and Rewriting Medicine
AlphaFold solved protein folding. AI is now designing drugs, decoding genomics, and accelerating clinical trials. This isn't incremental progress — it's a phase change in how we understand and manipulate biological systems. The upside is staggering: cures for diseases we've barely understood. The downside is equally sobering: who controls the biological playbook when it's written by machines?
19. Beyond the Button: The Death of the Interface and the Rise of Intent-Based Computing
The GUI — the graphical interface that has defined computing since 1984 — is becoming obsolete. When you can describe what you want in natural language and an AI executes it, the button, the menu, the dropdown, the form field all become unnecessary friction. We're moving from "point and click" to "state your intent." Every SaaS product built around complex interfaces is now sitting on a ticking clock.
20. The Panopticon Upgrade: AI-Powered Surveillance and the End of Anonymity
Facial recognition. Predictive policing. Real-time behavioral analysis at scale. AI doesn't just enable surveillance — it makes it comprehensive, automated, and cheap. Governments and corporations now have the tools to monitor populations at a granularity that was previously impossible. The question isn't whether this technology will be used. It's whether any meaningful privacy survives.
Closing
21. Post-Software Strategy: How to Build and Thrive in a World AI Has Already Eaten
The world AI is eating isn't coming — it's here. The companies, developers, and institutions that will thrive aren't the ones resisting the change or blindly embracing it. They're the ones who understand what's actually happening, make deliberate choices about where AI creates genuine value versus where it destroys it, and build with the awareness that the ground is still shifting. This is your field guide. Now go build accordingly.
This is a living document. AI is eating the world in real time, and the table of contents will grow with it.