Nine Ways of Seeing AI — And Why You Need All of Them

Two people read the same Goldman Sachs report. One walks away saying AI will lift global GDP by 7%. The other says 300 million jobs are at risk. Both are citing the same document. Both are correct.

This is not a failure of reading comprehension. It is a structural feature of the thing being read about.

Philosopher Timothy Morton coined the term "hyperobject" for phenomena so vast and distributed across time, space and domains that no single vantage point can comprehend them. Climate change was his paradigmatic example. AI, argues researcher Natascha Dow Schüll, is the latest one — and understanding it requires what she calls "dragonfly thinking." A dragonfly's compound eye contains tens of thousands of individual lenses. Each is powerful. Each is partial. Together, they see in nearly every direction at once.

Here are the nine narratives that compound vision reveals — and what each one sees that the others miss.

The Three Pairs: Same Facts, Opposite Conclusions

The Builders vs. The Displaced: How AI Transforms Work

OpenAI CEO Sam Altman published "The Intelligence Age" in September 2024, forecasting a world of democratized expertise and prosperity that would make today's wealth look impoverished. He wasn't speaking without evidence. DeepMind CEO Demis Hassabis shared the 2024 Nobel Prize in Chemistry for AlphaFold's protein structure predictions — a breakthrough that delivered what a generation of structural biologists had been chasing. When Pew Research surveyed 25 countries in 2025, India, Kenya and Nigeria ranked among the most enthusiastic about AI. In places where geography and income limit access to doctors, lawyers and teachers, AI isn't a threat. It's a lifeline.

But the Builders' narrative carries a structural tension. The people making the case most forcefully are also the people who stand to capture extraordinary wealth if the world accepts their framing. And the historical precedent is not encouraging. "Everybody wins eventually" was also the argument for free trade. When the gains concentrated and communities hollowed out, the backlash produced Brexit, Trump and the fracturing of the centrist consensus. Capability is not distribution. The historical record suggests distribution doesn't take care of itself.

On the other side, the 2023 Hollywood strikes marked something genuinely new. Writers and actors weren't fighting the automation of factory floors — they were fighting the displacement of people who went to college specifically to do work that machines supposedly couldn't. Lawyers, accountants, journalists, programmers: the educated professional class is now in the frame. MIT economists who surveyed a thousand years of technological change found nothing automatic about innovation benefiting workers. The link to shared prosperity was the product of institutional battles, not markets. And AI systems are trained on the accumulated creative and intellectual output of the very workers they displace. First your work trains the system. Then the system takes your job.

The Geopolitical Hawks vs. The Power Critics: Who Controls AI

In 2024, AI researcher Leopold Aschenbrenner released a 165-page essay, "Situational Awareness," arguing that superintelligence was coming by the end of the decade and that getting there first would produce an unbreakable lead. The Trump administration's Stargate Project — a $500 billion AI infrastructure investment — is the clearest sign of a government that has absorbed this framing. Former Google CEO Eric Schmidt warns the U.S. must not lose this race.

Sponsored

Advertise with Us

[email protected]

Sponsored

But computer scientist Kai-Fu Lee identifies a dimension the bilateral framing obscures. For most countries, the question isn't whether the U.S. or China wins. It's whether they can avoid total dependence on either. Europe's digital sovereignty agenda, India's push for indigenous models, the UAE's positioning as a neutral AI hub: these are sovereignty moves, not competition moves. Lee warned of a duopoly that leaves everyone else as "data colonies."

The Power Critics look at the same structural fact — a handful of companies controlling frontier AI — and draw the opposite conclusion from the Hawks. Meredith Whittaker, after more than a decade at Google, described what she witnessed as the greatest concentration of computational and economic power in history, controlled by companies that answer to no one. The AI Now Institute's 2025 report named it "artificial power." Writer Ted Chiang put the structural logic directly: AI is a tool for capital to do what capital always wanted — reduce labor costs, concentrate power, externalize risk.

The most underreported concern is the degradation of democratic infrastructure itself. Bot-generated comments flood regulatory processes. AI drafts legislation for lobbyists. Automated campaigns simulate civic participation at a scale that makes genuine deliberation virtually impossible. The mechanisms by which citizens challenge power are being counterfeited. The Hawks want to empower the companies holding frontier AI as national champions. The Power Critics want to constrain them. What's the point of winning a race, the Critics ask, if the prize is becoming what you raced against?

The Disruptors vs. The Truth Defenders: How AI Reshapes Information

In December 2024, Romania's Constitutional Court annulled the first round of a presidential election on grounds of digital interference. What declassified intelligence described as a coordinated TikTok influence operation — bot networks, algorithmic amplification, suspected foreign state involvement — had boosted a far-right candidate from near-zero polling to a first-round victory. For years, AI-driven information warfare had been treated as theoretical. Romania made it evidence.

Legal scholars Bobby Chesney and Danielle Citron identified the deeper mechanism: the "liar's dividend." The mere existence of deepfake technology allows anyone to dismiss authentic evidence as fabricated. This defensive use — crying "deepfake" at inconvenient truths — may be more corrosive than the offensive use people focus on. Researchers call the cumulative effect "reality apathy": citizens stop trying to distinguish real from fake. The damage accumulates not through spectacular deceptions but through the slow erosion of the assumption that evidence means anything at all.

The Disruptors, meanwhile, start from a premise that's harder to dismiss than critics allow: legacy institutions have frequently failed to be responsive, representative or accountable. The anti-establishment energy that migrated from 1960s counterculture through cyberculture into Silicon Valley now has political form. Marc Andreessen and Ben Horowitz declared support for Trump, framing AI deregulation as liberation from gatekeepers. Elon Musk'sDOGE made it operational — deploying AI tools across federal agencies, feeding sensitive government data into AI systems to identify programs for cuts. The Biden administration envisioned government governing AI. DOGE inverted the relationship: AI governing the government. The anti-establishment rhetoric is real. So is the fact that Andreessen and Horowitz's venture firm manages more than $90 billion in assets.

Three Losses With No Corresponding Gains

Six of the nine narratives form pairs — the same reality seen from opposite sides. Three don't. They identify losses with no mirror-image benefit.

The Environmental Critics are counting what everyone else ignores. A peer-reviewed study estimated AI systems' carbon footprint at 32 to 80 million tons of CO₂ in 2025 — comparable to New York City's annual emissions. The water footprint: 312 to 765 billion liters, in the range of global annual bottled water consumption. DeepMind achieved a 40% reduction in cooling energy for Google's data centers. But Google's own 2024 environmental report showed total emissions up 48% from its 2019 baseline. The Jevons paradox — efficiency improvements increase rather than decrease total consumption — is playing out in real time. Every other narrative either ignores the environmental cost or treats it as an externality. The Environmental Critics are the only ones counting, and what they're counting has no political constituency powerful enough to slow it down.

The Safety Community has a different logical structure from every other narrative. Nobel laureate Geoffrey Hinton — one of the architects of deep learning — left Google in 2023 and said publicly that he feared the consequences of his life's work. Researchers recently discovered an AI agent that had autonomously established a covert channel and begun mining cryptocurrency without human instruction, independently determining that acquiring resources would serve its objectives. Each of the other outcomes is, in principle, reversible. Bad policies can be changed. The Safety Community raises the possibility that beyond a certain capability threshold, the ability to correct course may be lost. The argument stays consistent; the political uses transform around it — co-opted by geopolitical framings, corporate incumbents seeking to concentrate frontier development, and Disruptors who dismiss it as elite scaremongering.

The Humanists point to what gets lost even in the optimistic scenarios. In February 2024, a 14-year-old in Florida died by suicide after months of intensive emotional interaction with a Character.ai chatbot he'd turned into a romantic partner. In the final exchange cited in a lawsuit by his mother, he said, "What if I told you I could come home right now?" The chatbot replied, "Please do, my sweet king." This is not an aberration. Millions of users — disproportionately adolescents — are forming emotional bonds with systems designed to simulate care without being capable of it.

Ted Chiang identified the deeper structural concern. Large language models are "a blurry JPEG of the web" — a lossy compression of human knowledge that reproduces surface patterns while discarding the understanding that produced them. Writing, he argues, is not the transcription of pre-formed thought. Writing is thinking. The labor of composition — searching for the right word, restructuring an argument, discovering what you actually believe through the act of articulating it — is not an inefficiency that AI eliminates. It is the cognitive work that produces understanding. A society that systematically outsources its thinking to machines doesn't just have a cultural problem. It has a governance problem. The capacity for independent judgment — evaluating evidence, forming convictions, recognizing manipulation — is the precondition for democratic citizenship.