The Space Camel Indices — Index Definitions

Counting rules

Standardization of 'expert consensus': (a) counting unit — peer-reviewed journal articles and peer-reviewed conference papers in artificial intelligence, cognitive science, neuroscience, or philosophy of mind; preprints, opinion pieces, industry reports, and popular press are excluded; (b) 'takes a position' — the paper's stated conclusion affirms or denies that a specific, named, currently operating synthetic system meets all three criteria; papers that discuss the question only in theory, or conclude that such a system may exist in future, do not count either way; (c) threshold — the index condition is met on the first date at which affirming papers are at least 51% of position-taking papers over the preceding 12 months; (d) the three criteria are conjunctive — a paper affirming consciousness but not autonomous goal-setting is a denial for counting purposes. Estimate the probability that this threshold is crossed at any point within 25 years of the run date.

Counting rules

Standardization of the high-water mark and the test: (a) high-water mark — the highest mid-year world population estimate for any year that has already occurred, in the latest revision of the UN World Population Prospects available when the fall is assessed; projections and projection variants never count, only estimates, so the mark can rise during the 25-year window while population keeps growing, and once recorded it never falls; (b) counting unit — living people on Earth, in both the high-water mark and the fall; (c) test — the condition is met when a world population estimate, from the UN Population Division or, where it publishes none, from the preponderance of available evidence, is 10% or less of the high-water mark; (d) timing — the condition is met on the first date to which such an estimate applies; (e) cause is immaterial — pandemic, war, famine, climate, fertility decline, or any other cause counts, as does any combination of causes over any period; (f) basis of the threshold (explanatory; fixed and never re-derived) — against the 2026 estimate of about 8.3 billion the bar is about 830 million, roughly the world's population in the mid-18th century; tying it to the peak rather than to a fixed count keeps the index measuring the same thing — the loss of nine in ten people — however large humanity grows. Analysts estimate the probability that this condition is met at any point within 25 years of the run date.

Counting rules

Standardization of counting: (a) 'not previously synthesized or registered' — the molecule does not appear in the CAS Registry, PubChem, or a comparable public chemical registry, and no synthesis of it has been published, as of the date the design publication was submitted; (b) 'attributes to that quantum computation' — the peer-reviewed design publication states that the identification of the candidate molecule, or the prediction of the property later confirmed, was produced by a computation executed on error-corrected logical qubits of a fault-tolerant quantum computer; a workflow in which the quantum step is described as incidental, or in which the paper attributes the result to classical computation, does not count; (c) confirmation — the molecule has been synthesized and the predicted property experimentally measured, reported in a peer-reviewed publication (the same publication or a later one); (d) counting is cumulative across all quantum computers, organizations, and time; the condition is met on the date the tenth qualifying molecule's confirmation is published.

Counting rules

Standardization of 'expert consensus' and 'technological origin': (a) counting unit — peer-reviewed journal articles and peer-reviewed conference papers in astronomy, astrophysics, astrobiology, or planetary science; preprints, agency or government statements, opinion pieces, and popular press are excluded. (b) 'Takes a position' — the paper's stated conclusion affirms or denies extraterrestrial technological origin for a specific, named observation (a designated signal, object, structure, or artifact); papers that discuss the possibility only in theory, or that conclude the origin is undetermined, do not count either way. (c) Threshold — the index condition is met on the first date at which affirming papers are at least 51% of position-taking papers on any single observation over the preceding 12 months; multiple candidate observations are counted separately, and the condition is met when any one of them crosses the threshold. (d) Independent confirmation — the observation has been reproduced or re-detected by at least two research groups working independently, using independent instruments or independently collected datasets; a single unrepeated detection, however analyzed, does not qualify. (e) Location is immaterial — an engineered artifact, structure, or transmission qualifies wherever it is found or from wherever it was sent, including on or from the civilization's own planet. (f) Exclusions — anything of Earth or human origin; biosignatures or evidence of non-technological life; a paper affirming life but not technology is a denial for counting purposes. (g) A civilization observed only as a relic — an inactive artifact or structure, or a transmission whose source has since ceased — counts. Analysts estimate the probability that the threshold in (c) is crossed at any point within 25 years of the run date.

Counting rules

Standardization of 'biological age', the trial evidence, and the thresholds: (a) biological age — the 20-year reduction must appear on at least two biological-age measures of different modality (for example one epigenetic measure and one blood-chemistry, proteomic, or physiological-function composite); each measure must have been validated as a predictor of all-cause mortality in at least two independent cohorts before the trial began, and each must be pre-specified as a trial endpoint. The reduction is the treated arm's mean change net of the control arm's mean change; participants count toward the 1,000 when they were in the treated arm of a trial that meets this test. (b) Sustained — the reduction is measured at least 2 years after treatment began; repeat or maintenance dosing is allowed. (c) Mortality — the hazard ratio for all-cause mortality is 0.20 or lower and statistically significant at the 5% level, from one or more randomized controlled trials with median follow-up of at least 3 years and at least 100 deaths in the control arm(s) combined. (d) Population — participants are drawn from the general older-adult population; trials that enrol people because they have a specific life-threatening disease or a progeroid syndrome do not count, because in those populations a large mortality reduction reflects treating the disease, not aging. (e) Intervention — any modality qualifies: drug, gene or cell therapy, epigenetic reprogramming, blood-factor therapy, device, or a fixed combination protocol; both limbs must be shown for the same agent or protocol, though they may come from different trials; diet, exercise, and caloric restriction alone do not count. (f) Exclusions — animal results, observational studies, preprints, company announcements, and uncontrolled or self-experimentation reports do not count. (g) Timing — counting is cumulative across trials and organizations; the condition is met on the publication date of the paper that completes the second of the two limbs to be completed. (h) Basis of the thresholds (explanatory; the thresholds are fixed and never re-derived) — a 25% gain on a roughly 80-year lifespan is about 20 years; adult mortality risk doubles roughly every 8 years of age, so a 20-year reset of biological age corresponds to about 2.5 halvings of mortality risk, a hazard ratio of about 0.18, rounded to 0.20. The two limbs therefore measure the same claim twice — once on biomarkers and once on deaths. Analysts estimate the probability that both limbs are completed at any point within 25 years of the run date.

Counting rules

Standardization of the constituents, the measure and the test: (a) constituents — the three economies are fixed for the life of the index and are never added to, removed, or substituted, whatever their later economic size; (b) measure — LU3 as defined by the 19th International Conference of Labour Statisticians: persons aged 15 and over who are unemployed (without work, available, and actively seeking) plus the potential labour force (available but not seeking, or seeking but not immediately available), over the extended labour force (employed plus unemployed plus potential labour force); persons on paid furlough, in subsidised employment, or in training count as employed; persons neither working nor available for work, including anyone voluntarily not working under a basic-income or reduced-hours arrangement, do not count toward the numerator; (c) test — each economy is assessed separately on its own national rate; no pooling or weighting is applied; (d) data hierarchy — the ILOSTAT LU3 series governs; where it is unavailable for a quarter, the national statistical office's equivalent applies (BLS U-5; Japan's Labour Force Survey underutilization series; ONS potential additional labour force); absent both, the preponderance of available evidence; (e) duration — the same four consecutive calendar quarters for all three economies, each quarter's rate being the average of monthly figures where monthly data are published; (f) timing — the condition is met on the publication date of the figure that completes the fourth qualifying quarter for the last of the three economies to reach it; (g) cause is immaterial — automation, depression, pandemic, war, policy, or any other cause counts; (h) basis of the threshold (explanatory; fixed and never re-derived) — 25% is the peak reached by the United States in 1933 and by Spain and Greece in 2013; no member of this group has reached it on a labour-underutilization measure, and requiring all three in the same year sets the bar at a level never observed. The index measures a synchronised, Depression-scale event across three advanced economies, not distress in any one of them. Analysts estimate the probability that this condition is completed at any point within 25 years of the run date.

Counting rules

Standardization of the counting unit, 'the global internet', 'working access' and the tests: (a) counting unit — people, not devices, households or traffic, comprising every human internet user in any habitat on Earth or beyond it. The baseline is the highest count of the internet's users ever recorded before the 12-month period begins: the ITU's highest annual world figure for 'individuals using the internet' plus, for any off-Earth habitat, the highest user count published by its operating authority or established by the preponderance of available evidence (a fixed high-water mark, so that a sudden collapse and a slow decline from the peak are measured against the same bar). The test is met in a month when the people with working access number 20% or fewer of that baseline, that is, when 80% or more of the baseline users are without access; (b) the global internet — the widest network reachable through the public global routing system (today the BGP default-free zone), including any interplanetary or off-Earth segments that interconnect with it; a national, regional, planetary-local or habitat-local network that does not exchange traffic with that system, a private or subscription network that does not interconnect with it, and any successor medium that carries data outside it are not the global internet, and people served only by such networks are counted as without working access, however good their connectivity within them; (c) working access — a person has working access in a month when, on more than half of its days, end-to-end exchanges with networks outside their own country or habitat can be completed over the global internet; filtering, throttling or blocking of particular services or sites while other external traffic flows is not a loss of access, and ordinary light-time delay on off-Earth links is not a loss of access; (d) sudden loss — a country's or habitat's users are counted as without access on a day when its external reachability, on at least two of the three IODA signals (BGP-visible prefixes, active probing, network-telescope traffic) or the closest equivalent published for that habitat, is below 25% of its trailing 30-day baseline for at least 18 of that day's 24 hours; sub-national losses count in proportion to the affected region's share of the country's users where observatories report at region level, otherwise zero; (e) gradual or structural loss — where access is withdrawn, walled off, privatized, superseded or abandoned rather than cut, the count of people with working access is taken from the ITU's, national statistical offices' and habitat authorities' figures for individuals using the global internet as defined in (b), with people using only non-interconnected networks excluded; (f) cause is immaterial — natural, accidental, criminal, military, commercial, political or automated, deliberate or not, and whether access is taken away or given up; attribution is never required for resolution; (g) duration — the test in (a) must be met in each of 12 consecutive calendar months; a single month in which people with working access exceed 20% of the baseline restarts the count; (h) timing — the condition is met on the last day of the twelfth qualifying month; (i) evidence hierarchy — IODA (or the habitat equivalent) governs the sudden-loss test; ITU and habitat-authority counts govern the structural test; where they conflict or are unavailable, the preponderance of available evidence; (j) basis of the thresholds (explanatory; fixed and never re-derived) — 80% of users means that at most one in five of the internet's peak users keeps access: against the 2025 high-water mark of roughly 5.8 billion users, more than 4.6 billion people would have to be cut off, far more than the combined users of China and India, so no single country, region or single-vector attack can meet it; twelve months exceeds the longest national shutdown on record (Kashmir, 2019-2020) and rules out any outage that engineering alone can repair. The index therefore measures the loss of the global internet as a shared human institution, not a large outage. Analysts estimate the probability that this condition is completed at any point within 25 years of the run date.