Battle of Ideas

Zoos should be banned

A steelman map of the public fight over zoos — arguments for and against, with sources and assumptions.

AI-generated · paired steelman agents · independently red-teamed · Pass-1 source spot-checks only · framing-fidelity not independently verified · single model family

Keeping wild animals on public display (zoos, menageries; aquaria in the same fight). Sanctuaries and breeding without a paying audience are a third camp. Farm animals and pets are not the claim here.

AGAINST 5

no further strong arguments at this depth

FOR 5

no further strong arguments at this depth

People also ask

Questions people actually type. The two columns above are the cases — open a card for sources, assumptions, and counters.

Should zoos be banned?

Zoos should be banned pros and cons

Should zoos be banned or not?

Zoo animal welfare

Reasons zoos should be banned

Zoos should be banned worldwide

Ordering within each column: strongest first — validation tier, then source quality, then representativeness.

AGAINST · Zoos should be banned
Empirical — moderateP1

A visit is the only mass contact most people will ever have with a wild animal that is not a screen

Moss, Jensen and Gusset (Conservation Biology, 2015; Nature, 2014) is FOR's 'thin quiz' paper. Read the other way: 5,661 visitors to 26 institutions in 19 countries, the largest international zoo-and-aquarium education study then on offer, found aggregate biodiversity understanding moving 69.8% to 75.1% and named individual actions 50.5% to 58.8% in one visit. The 2017 campaign paper found a larger bump when people actually saw the materials. WAZA's 700 million visits is the multiplier. AGAINST does not need the quiz to be a biology degree. It needs the fact that most humans will never stand in a condor release site or a Mongolian steppe. A living animal in a well-run house is a different object from a clip. Clubb and Mason's ranging critique can still be true for polar bears and false as a reason to zero the contact. Aquaria are in the claim: a public tank that is not an orca show is how inland children ever see a living reef animal. Ban the worst tanks. That is a collection policy. It is not 'zoos should be banned.'

Key assumptions

  • A short-run knowledge bump, times hundreds of millions of visits, is a conservation-relevant education effect partial
  • Screens and school films are not a close substitute for seeing a living animal untestable

Red team — the strongest counters

A six-point quiz bump among people who already bought a ticket is not mass education

The sample is visitors, not the public. Immediate post-visit self-report is cheap. FOR's bar — does this pay for a life in an enclosure? — is the right bar, and a 5% shift does not obviously clear it.

'Not an orca' is how AGAINST quietly drops aquaria

The claim put aquaria in. If the steelman is 'ban the worst tanks,' that is not AGAINST banning zoos. It is a third-camp collection policy FOR already offered.

Sources

Confidence, decomposed

Logical validity●●●●○
Premise support●●●○○
Representativeness●●●●○
Source quality●●●●●

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
AGAINST · Zoos should be banned
Empirical — moderateP1

Some species only exist because zoos bred them when the wild was empty

US Fish and Wildlife's California Condor Recovery Program, 31 December 2025: 607 birds in the world, 392 wild, 215 in human care. Captive breeding is still at Los Angeles Zoo, San Diego Zoo and Safari Park, Oregon Zoo, the Peregrine Fund, and Chapultepec. In 1987 the last wild birds were taken in; San Diego hatched the first captive chick in 1988; releases began in 1992. The wild flock is still fed by captive-bred releases (46 in 2025). Prague Zoo's Przewalski's horse programme is the other named specimen: the species died out in the wild, survived in zoo studbooks (Prague has kept the international studbook since 1959), and has been flown back to Mongolia and, more recently, Kazakhstan; National Geographic's 2025 account puts more than a thousand in China and Mongolia. AGAINST does not need every zoo giraffe to be a condor. It needs existence proof that public-display institutions did the breeding that a 'sanctuary without an audience' had not yet done, at the moment the wild stock hit zero. A ban written as 'keep the hospital, kill the zoo' has to say who runs the hospital tomorrow, with which birds, on which budget.

Key assumptions

  • The condor and Przewalski programmes required zoo husbandry capacity, not only a government aviary that could have existed without display partial
  • Enough other threatened species still need that capacity that a display ban would cost recoveries, not only Saturday photos partial

Red team — the strongest counters

Two flagship recoveries do not licence a million display animals

Condors and Przewalski's horses are the brochure. Most zoo occupants are not in a release pipeline. You can keep those two programmes as closed breeding centres — the third camp the scope already named — without a paying giraffe. Existence proof of an ark is not existence proof that the menagerie must stay open.

The wild is still the problem — lead, habitat, hunting — not a shortage of cages

Condor recovery still depends on stopping lead ammunition. Zoo chicks dumped into a poisoned range are a subsidy for a policy the zoo does not control. Breeding is not conservation if release is a revolving door.

Sources

Confidence, decomposed

Logical validity●●●●●
Premise support●●●●○
Representativeness●●●○○
Source quality●●●●●

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
AGAINST · Zoos should be banned
Empirical — moderateP1

The conservation budget is small next to tickets — and still one of the larger private conservation budgets on earth

WAZA's public line: more than 700 million visits a year, and zoos and aquariums as 'the third largest funder of conservation globally,' with the 2015 Conservation Strategy's 'over 350 million US dollars' a year on wildlife conservation (from a survey of associations, almost certainly an undercount by WAZA's own note). AZA members: $341.1 million on field conservation in 2024, 397 IUCN-threatened species or subspecies listed as beneficiaries, plus $40.7 million on AZA SAFE programmes. Keulartz (Animals, 2023) repeats the third-largest-funder claim after two larger public sources. FOR's ratio — $341 million versus $35 billion of economic impact — is true and does not make $341 million trivial in the conservation world, where many field projects are tens of thousands. AGAINST's claim: the visitor business is how those projects are funded in this institutional form. Sanctuaries-without-display do not currently write that cheque at that scale. A ban is a bet that governments or donors will. That is a funding hope, not a finding that the animals on show did no conservation work.

Key assumptions

  • A material share of that field spend would not be replaced, at similar scale, if display were banned partial
  • WAZA/AZA self-reported conservation dollars are close enough to real field spend to matter partial

Red team — the strongest counters

Self-reported 'conservation spend' can be a branding line-item

What counts as field conservation is the association's definition. A keeper's salary in a city, a sign about plastic, and a cheque to a reserve are not the same dollar. Third-largest funder is a slogan until audited.

States already fund conservation without selling pacing bears

National parks, environment ministries, and NGOs exist. If the $341 million is real, it is a political choice to attach it to a zoo rather than a tax. FOR's 'replacement appropriation' is exactly what governments do for other public goods.

Sources

Confidence, decomposed

Logical validity●●●●○
Premise support●●●○○
Representativeness●●●●○
Source quality●●●●○

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
AGAINST · Zoos should be banned
Logically validP1

Clubb and Mason did not say 'ban zoos' — they said stop keeping the species that cannot cope

Clubb and Mason (Nature, 2003) wrote that naturally wide-ranging carnivores should be 'fundamentally improved or phased out.' They also wrote that some species 'apparently thrive in captivity' — their examples were ring-tailed lemurs and snow leopards — and that the unexplained variation in welfare was what they were trying to predict. The 2007 follow-up is even more operational: enrich for ranging, or stop keeping the high-range species, or concentrate breeding on species that fit current enclosure sizes. AGAINST takes that as the honest welfare programme: shrink the mismatched stock, enlarge the houses that remain, measure stereotypy, publish cub mortality. A zoo that is only condors, small-range ungulates, and a decent aquarium is still a zoo in the claim's sense — public display of wild animals — and is not the polar-bear-pacing poster. FOR's jump from 'this species should not be here' to 'the institution should be illegal' is the smash. Pets and farm animals are out of scope. So is a sanctuary. What remains is display. Display of animals that cope is the part Clubb and Mason did not indict.

Key assumptions

  • A large enough share of current collections can be the 'thrive' class, or moved into it, that a reformed zoo is still a zoo rather than an empty brand partial
  • Phase-out of wide-ranging carnivores is politically and financially available inside the display model partial

Red team — the strongest counters

The posters on the gate are the high-range species

People do not queue for a snow leopard they cannot see and a ring-tailed lemur. The business model FOR named — Saturday tickets — is the polar bear and the orca. A 'reformed zoo' that phases those out may not have a gate. Then you have the third camp anyway.

2003 'thrive' is not a welfare pass in 2026

Elephants, great apes, and many birds have accumulated welfare cases Clubb and Mason did not put in the carnivore table. Using two happy examples as a licence for public display is the industry's oldest move.

Sources

  • Captivity effects on wide-ranging carnivores Clubb & Mason, Nature 2003;425:473–474. Thrive examples (ring-tailed lemurs, snow leopards) vs problems (Asian elephants, polar bears); phase out or improve the wide-ranging. Pass-1: already checked. AGAINST uses the paper's own split, not a later industry brochure. P1 checked
  • Natural behavioural biology as a risk factor in carnivore welfare Clubb & Mason, Appl Anim Behav Sci 2007. Operational: more space / variability, or phase out high-range species in favour of small-range ones. Pass-1: already checked. P1 checked

Confidence, decomposed

Logical validity●●●●●
Premise support●●●●○
Representativeness●●●●○
Source quality●●●●●

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
AGAINST · Zoos should be banned
Logically validP1

The wild is not a kindness machine, and a ban does not create a spare continent

The third camp — sanctuaries and breeding without a paying audience — is in the scope so that neither column has to pretend every captive animal can go 'home.' AGAINST's use of it: capacity is the constraint. AZA's accredited members already hold the animals they hold; WAZA's 700 million visits are how those houses are paid for. If display is banned, the animals do not teleport to a steppe. They wait in the same buildings with less money, or they are moved into a smaller number of closed facilities, or they are killed. Condor and Przewalski programmes show release is possible for a few species with habitat and a recovery plan. They do not show a spare range for a city zoo's tigers, sea lions, and parrots. FOR's 'keep the hospital' needs hospital beds. Until those exist, a ban is a welfare slogan over a logistics hole. Aquaria in: a reef tank is not an orca, and dumping it does not refill a reef. The honest sequence is phase out the Clubb-and-Mason disasters, fund the arks, then argue about the rest of the gate. A worldwide ban as the first move skips the sequence.

Key assumptions

  • Sanctuary and closed-breeding capacity cannot absorb current display collections on a ban-relevant timescale partial
  • Killing or warehousing surplus animals would be worse, in welfare terms, than continued reformed display untestable

Red team — the strongest counters

Capacity that depends on tickets is the problem, not a constraint we must obey

Every industry says 'you cannot close us, think of the stock.' Breeding less, stopping imports, and letting collections shrink is the phase-out. Treating today's headcount as sacred is how display never ends.

A cull of surplus is already zoo practice — it is not a unique horror of a ban

European zoos already manage 'surplus' animals. If the welfare hole is killing, the industry is in it. Using sanctuary scarcity as a reason to keep the Saturday show is a hostage argument.

Sources

Confidence, decomposed

Logical validity●●●●●
Premise support●●●○○
Representativeness●●●●○
Source quality●●●●○

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
FOR · Zoos should be banned
Empirical — moderateP1

The animals people come to see are the ones that fare worst in a cage

Clubb and Mason (Nature, 2003) collated captive welfare against wild ranging in carnivores: natural home-range size predicted both stereotypic pacing and captive infant mortality, and the pattern was not explained by body weight or by husbandry variation they could measure. Their conclusion was not a brochure tweak. It was that keeping naturally wide-ranging carnivores should be 'fundamentally improved or phased out.' Secondary reporting of the same paper (and the authors' own follow-up in Applied Animal Behaviour Science, 2007) puts a polar bear's typical zoo enclosure on the order of a millionth of a wild range, with pacing eating a large share of some animals' day and high captive cub death. Snow leopards and small-range species can look fine; those are not the posters on the gate. FOR's load-bearing move: the public-display business is built on the species whose biology Clubb and Mason said does not fit the box. Enrichment and a bigger moat are the industry's reply. They are not a second home range. Farm animals and pets are not this claim. Wild animals on show are.

Key assumptions

  • Stereotypy and extra infant mortality are welfare harms serious enough to outweigh whatever the visitor learns untestable
  • Headline zoo species remain, in current enclosures, in Clubb and Mason's high-range class testable

Red team — the strongest counters

2003 ranging data are not 2026 enclosures

Clubb and Mason's dataset closed in the late 1990s. Mega-exhibits, freeze-thaw ice, and rotational yards exist now. A polar bear still does not get Nunavut, but 'one-millionth' is an old typical, not a proof that every current house fails. Phasing out the worst species is a collection policy, not a ban on zoos.

Stereotypy is not a one-to-one welfare meter

Mason and Latham have themselves argued that some stereotypies can be scars or coping, not a live scream. Infant mortality in captivity can reflect veterinary recording and hand-rearing rules as much as misery. The Nature sentence 'phase out' is a recommendation from two Oxford biologists, not a finding that display itself is torture.

Sources

  • Captivity effects on wide-ranging carnivores Clubb & Mason, Nature 2003;425:473–474. Wild home-range size predicts captive stereotypy and infant mortality; phase out or fundamentally improve wide-ranging carnivores. Pass-1: Nature page exists (doi 10.1038/425473a; PMID 14523435). P1 checked
  • Natural behavioural biology as a risk factor in carnivore welfare Clubb & Mason, Applied Animal Behaviour Science 2007;102:303–328. Same comparative result in more detail; ranging, not hunting, predicted stereotypy. Pass-1: journal record exists (doi 10.1016/j.applanim.2006.05.033). P1 checked

Confidence, decomposed

Logical validity●●●●●
Premise support●●●●○
Representativeness●●●●○
Source quality●●●●●

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
FOR · Zoos should be banned
Empirical — moderateP1

A five-point bump on a biodiversity quiz does not pay for a life in an enclosure

Moss, Jensen and Gusset (Conservation Biology, 2015; Nature correspondence, 2014) ran the largest zoo-and-aquarium visitor study WAZA could boast: thousands of visitors to dozens of institutions, pre- and post-visit. Aggregate 'some positive evidence of biodiversity understanding' moved from 69.8% to 75.1%; the share who could name an individual action to help biodiversity moved from 50.5% to 58.8%. A later campaign evaluation (Frontiers in Ecology and the Environment, 2017) found similar pre/post bumps, larger when people reported seeing the campaign panels. FOR does not need to call the researchers liars. A six-point shift on a survey, among people who already chose to walk through a zoo, is a thin educational yield. It is also self-report immediately after the gift shop. The Clubb and Mason animal is still pacing. If education is the moral defence of display, this is the measured education. It does not clear the bar. School films, field reserves, and sanctuaries that do not sell a spectacle are still available. They are the third camp, not a zoo with better signage.

Key assumptions

  • A short-run survey bump is a fair measure of the education zoos claim, or at least the best they have published partial
  • Even a larger, longer-lasting knowledge gain would not justify display if the welfare cost is of Clubb-and-Mason size untestable

Red team — the strongest counters

Pre/post on the same day is not 'thin' if you have no other mass contact with wild animals

Most people will never go to a field site. A 5–8 point shift across millions of visits is a lot of people. Jensen, Moss and Gusset's delayed follow-up even found action-knowledge still up years later (small n). FOR is demanding a randomised education miracle that no museum would survive either.

The quiz is not the whole education — children seeing a living animal is the point

A Likert bump understates awe, identity, and later donations. If FOR only counts survey items, it has defined education as the one thing zoos barely move, then declared the case closed.

Sources

Confidence, decomposed

Logical validity●●●●●
Premise support●●●●○
Representativeness●●●●○
Source quality●●●●●

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
FOR · Zoos should be banned
Empirical — moderateP1

The business is tickets — conservation is a side budget

WAZA's own home page and Conservation Strategy still advertise more than 700 million visits a year to the world's zoos and aquariums — 'one-tenth of the world's population' in the slogan — and a conservation spend on the order of US$350 million in the older global survey (the Strategy's 'over 350 million US dollars' line). AZA's 2024 member survey: $341.1 million on field conservation, and $35 billion in economic impact with more than 209 million visits. FOR can grant every dollar of the $341 million. The ratio is the argument. The institution that must fill a car park on a Saturday is an entertainment business with a conservation department, not a conservation organisation that happens to sell ice cream. Keulartz (Animals, 2023) repeats the 'third largest financial supporter of species and wildlife conservation' line for WAZA members after two larger public sources. Third-largest donor is compatible with the gate still being the point. Sanctuaries and breeding without a paying audience are the named third camp. This column is the paying display.

Key assumptions

  • If the visitor business closed, the conservation dollars would not mostly vanish with it — or that replacing those dollars is still better than keeping the cages partial
  • Economic-impact billions measure the display business, not a hidden conservation surplus partial

Red team — the strongest counters

The conservation budget is paid for by the tickets

If you ban the display, the $341 million is not sitting in a treasury waiting to be reallocated. It is a by-product of Saturday visits. 'Side budget' can be the only budget some endangered-species field projects have. FOR needs a replacement donor, not a ratio.

WAZA's 700 million is a marketing census, not a welfare finding

Visit counts include aquarium school trips and city parks with a few ungulates. Using 'entertainment' as a smear does not show the animals are harmed. A museum also sells tickets. The harm has to come from the other args.

Sources

  • WAZA — zoos and aquariums as conservation funders / 700 million visits waza.org: >700 million visits; 'third largest funder of conservation globally' claim. Conservation Strategy (2015) : over US$350 million spent on wildlife conservation per year (survey). Pass-1: WAZA pages exist. Treat $350M as WAZA's own figure, not an audited global ledger. P1 checked
  • AZA Zoo and Aquarium Statistics AZA: $341.1 million field conservation in 2024; $35 billion economic impact; >209 million visits. Pass-1: aza.org statistics page exists (updated 2026). US accredited members, not the world — a specimen of the ratio. P1 checked

Confidence, decomposed

Logical validity●●●●○
Premise support●●●●○
Representativeness●●●●○
Source quality●●●●○

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
FOR · Zoos should be banned
Empirical — moderateP1

Tanks are the same fight — an orca is not a goldfish

The scope puts aquaria in the same fight as zoos and menageries. Clubb and Mason's ranging logic does not stop at the tide line. Marino, Rose, Visser and colleagues (Journal of Veterinary Behavior, 2019) review captive orcas as a chronic-stress case: abnormal behaviour, social structures that would not exist in the ocean, and survival-to-milestone figures they read as worse than wild populations. Jett and Ventre (Marine Mammal Science, 2015) reported median survival in US parks above non-US parks, and better after 1985 than before — and still a picture in which most captive orcas do not see the ages wild females in the Pacific Northwest routinely do. SeaWorld-linked work (Robeck et al., 2015) disputes the life-expectancy gap. FOR should not pretend that dispute vanished. It should say: even if the latest-generation tank matches a wild mean age, the animal is still a pelagic hunter in a pool, on a show schedule, for a ticket. That is display of a wild animal. A ban that left the orca and the beluga in the building would not be this claim. Goldfish and farmed salmon are not. The wide-ranging mammal in the tank is.

Key assumptions

  • Cetacean welfare in tanks is bad enough that the remaining lifespan dispute does not rescue the display partial
  • The same display-versus-rescue split used for terrestrial zoos can be written for marine mammals partial

Red team — the strongest counters

The lifespan table is an active fight, and FOR picked the advocate column

Robeck, Willis, Jackman and colleagues estimated captive life expectancy much closer to wild once early decades are dropped. Marino's 2019 paper is a narrative review from the sanctuary movement. Using it as if the tank were settled science is the sympathetic register.

Most aquarium animals are not orcas

A ban aimed at killer whales is not a ban on public aquaria of fish, coral, and rays. FOR's 'tanks are the same fight' either overreaches to goldfish or quietly retreats to marine mammals. The claim said aquaria in. It did not say every tank is an orca tank.

Sources

  • The Harmful Effects of Captivity and Chronic Stress on the Well-being of Orcas (Orcinus orca) Marino, Rose, Visser, Rally, Ferdowsian & Slootsky, Journal of Veterinary Behavior 2019. Chronic-stress / abnormal-behaviour case against tanks. Pass-1: doi 10.1016/j.jveb.2019.05.005 exists. Authors are welfare advocates — empirical-moderate, not a randomised tank trial. P1 checked
  • Captive killer whale (Orcinus orca) survival Jett & Ventre, Marine Mammal Science 2015. Median survival higher in US and post-1985 cohorts, still poor versus wild female milestones in the authors' reading. Pass-1: journal record exists. Robeck et al. 2015 contest the life-expectancy comparison — FOR must not treat Jett as the last word on years. P1 checked

Confidence, decomposed

Logical validity●●●●○
Premise support●●●○○
Representativeness●●●●○
Source quality●●●●○

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.
FOR · Zoos should be banned
Logically validP1

A sanctuary without a paying audience is not a zoo — do not smuggle it into the ban

The scope already names the third camp: sanctuaries and breeding without a paying audience. FOR's job is not to pretend every captive wild animal can be shipped to a fictional wild. It is to stop making captivity a Saturday outing. A condor or ferret breeding programme that the public never sees is not a zoo in the sense this claim names. A city menagerie whose budget requires a giraffe you can photograph is. The industry's steelman is that the giraffe pays for the condor. That is the previous argument's ratio, and it is a contingent fundraising claim, not a law of nature. States already fund conservation without selling pacing carnivores. Private sanctuaries already take animals that cannot be released, on the rule that the visitor is not the product. Aquaria sit in the same split: a public tank of a wide-ranging pelagic is display; a rescue pool for a stranded seal is not. Ban the first. Do not pretend you have thereby banned the second.

Key assumptions

  • Sanctuaries and closed breeding centres can be funded and regulated without ticketed display as the engine partial
  • A legal ban on public display can be written so that bona fide rescue is not swept in partial

Red team — the strongest counters

Without visitors, the sanctuary becomes a warehouse you cannot inspect

Public opening is how outsiders see the animals. A closed breeding centre can hide bad husbandry as easily as a zoo can. 'Third camp' is clean on paper. The live risk is a ban that moves animals into worse, quieter cages.

The condor paid for by the giraffe is not a brochure — it is how several recovery programmes actually ran

If display money bought the spare birds, splitting 'hospital' from 'show' is an accounting wish. FOR needs the replacement appropriation before the ban, or the third camp dies with the zoo.

Sources

Confidence, decomposed

Logical validity●●●●●
Premise support●●●○○
Representativeness●●●●●
Source quality●●●●○

Provenance

Generated by a paired steelman agent (single model family) · red-teamed by an independent adversarial agent · sources Pass-1 spot-checked (existence and rough fit) — framing-fidelity not independently verified. Judged on merit: per the founding rule of this project, AI authorship is disclosed at site level and arguments stand or fall on their content.