The Law Says One Thing. Reality Rewards Another
Personal Perspective: Corporate power exploits human bias as much as policy.
Posted May 13, 2026 | Reviewed by Gary Drevitch
Historically a white elephant was a gift so expensive and useless that the recipient could not get rid of it — a burden maintained out of embarrassment rather than logic. The phrase has migrated. Today it refers to issues that are uncomfortable to talk about, and hence easier to shove underneath the verbal table than to debate and eliminate. It is a sadly fitting label for entire political architectures: systems of concentrated privilege that operate in full public view; topics that are documented by international bodies, reported by mainstream media, and debated in parliaments — and yet persist without serious correction.
Fossil fuel subsidies and artificial intelligence lobbying are among today's most instructive white elephants. They share a political grammar: a handful of firms become systemically important; restraint gets framed as national weakness; costs are smoothly offloaded onto citizens, workers, and ecosystems. Little of this is hidden. Most of it is condoned. By us.
Hardly hidden. Largely condoned
Condoning the offensive action of some is an act that requires the tacit acceptance of many. In the context of today’s hybrid world, a place in rapid planetary decline , it is as important to understand why ordinary people let costly absurdities stand, as is cataloguing the absurdities themselves. Our mind is not well-suited to trillion-dollar abstractions. It gravitates toward visible threats, immediate rewards, and familiar arrangements. Status quo bias refers to our tendency to favour existing systems; we register the sting of perceived loss roughly twice as sharply as the pleasure of equivalent gain — loss aversion strikes. Industries that profit from inaction have studied both. Every reform gets framed as a threat; every subsidy, a lifeline; every regulation, a job killer. The framing works because our evolutionary wiring renders us inclined toward it. Public condoning is the predictable output of cognitive tendencies being deliberately triggered. Two examples that may sound familiar:
Fossil fuels: The Law says transition, the budget says addiction
The numbers are stunning. According to the IMF's 2025 fossil fuel subsidies update , implicit subsidies — the portion that lets industrial fossil fuel users avoid paying for climate damage, air pollution, and forgone tax revenues — reached USD 6.7 trillion in 2024, equal to 5.8 percent of global GDP. Add explicit price support and the total exceeds USD 7.4 trillion. Governments are making fossil fuel subsidies appear cheaper than they cost by ensuring the damage settles elsewhere — on public health systems, on future generations, on anyone without the lobbying budget to push back.
Article 2.1(c) of the Paris Agreement calls for making finance flows consistent with a pathway toward low greenhouse gas emissions. That is plain language. The International Energy Agency has stated that in a net-zero pathway, any new oil and gas resource developments would need to be matched by production declines elsewhere to avoid lock-in. Legal obligation and public spending point in opposite directions. Both facts are publicly available. Neither triggers large scale public outcrys, or policy correction.
This is where cognitive bias does its political magic. Climate damage arrives slowly and diffusely; subsidies arrive quickly, in budget line items. The availability heuristic — the human tendency to weigh visible, concrete information more heavily than distant, statistical evidence — makes it genuinely difficult for people to hold both realities with equal urgency. When pollution gets rebranded "energy security" and clean energy postponement gets sold as "pragmatism," the rebrand succeeds because our internal pattern-matching system is already searching for those anchors. The machinery of delay does not need to lie outright. It only needs to feed the bias that was already there.
AI: the new infrastructure of influence
Artificial intelligence is sold as a fresh chapter. The political architecture is recognisable. The OECD has noted that competition in generative AI faces risks from barriers to key inputs — high-quality data, computing power, and capital. The three largest cloud providers now hold over 60 percent of the global market. Whoever controls chips, cloud infrastructure, and investment pipelines determines who gets to build, who gets to scale, and who gets squeezed out before the public conversation has even formed.
Then comes the lobbying — visible, documented, and growing. Issue One found that seven major tech and AI companies spent a combined USD 50 million on US federal lobbying in just the first nine months of 2025 — roughly USD 400,000 for every day Congress was in session. By full-year 2025, the sector had exceeded USD 100 million for the first time. The pace accelerated into 2026: in Q1 alone, eleven leading tech companies spent USD 20 million — USD 226,000 a day. None of this is a secret. It is filed, published, and freely available. What is scarce is public response.
The condoning instinct surfaced vividly in 2025, when a provision to block U.S. states from regulating AI for a decade was defeated in the Senate 99 to 1. What matters as much as the defeat is what the proposal revealed: it was advanced while AI harms and liability remain legally unsettled. Build first, lobby hard, ask the Law to wait. The Senate stopped it this time. The spending that produced it continues unchanged.
A further cognitive risk compounds the political one. Automation bias refers to the human tendency to defer to machine outputs as more authoritative than human judgment. As AI embeds itself in hiring, lending, healthcare triage, and public services, these systems accumulate social trust faster than accountability structures can form. Concentrated infrastructure, deferred regulation, and popular deference to machine authority compose a combustible combination — and all three elements are advancing simultaneously, in plain sight.
The same playbook, different branding
The parallel between fossil fuels and AI runs through political mechanics. A handful of firms become systemically important. Restraint gets framed as dangerous — bad for jobs, competitiveness, national strength. Costs get externalised. In the fossil case: pollution and climate damage absorbed by public health systems and future generations. In the AI case: concentrated power, weak accountability, copyright conflict, and rising energy and water demand. UNEP has already warned that data centre energy and water consumption is accelerating, driven in part by AI expansion. The externalities arrive before the guardrails — and the pattern, by now, is familiar enough to be anticipated.
What keeps the playbook working is collective tolerance. Condoning is an easier default than confronting. The status quo feels stable even when its costs are enormous. Loss aversion makes reform feel riskier than continuation. And availability bias ensures that abstract future damage always loses the competition for attention against concrete present spending. White elephants do not survive through concealment. They survive through exhaustion and normalisation.
Four anchors for agency
The law can say transition, rights and accountability. Reality will keep moving in the opposite direction whenever money, infrastructure and behavioral capture overrule public purpose. The task now is to stop treating the contradiction as normal.
Share this post Facebook Bluesky Linkedin Email
There was a problem adding your email address. Please try again.
By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy
Cornelia C. Walther, Ph.D., is an Associate Professor at Sunway University and a Wharton/University of Pennsylvania Fellow who researches hybrid intelligence and ProSocial Al.
Get the help you need from a therapist near you–a FREE service from Psychology Today.
This article is part of the Bringwise Psychology Journal — daily insights on human behavior, mental health, and personal growth.