Nostradamus Predictions for 2029: The Computed Reading
No quatrain names 2029 — and 2029 is the year this ledger is graded at scale. Three sealed anchors resolve by December of that year: record US debt (78%), no US–China live fire (82%), majority weekly AI use (62%). This dossier shows exactly how the reading is computed — base rates, era resonance, the Loom's couplings, and the seals that pin it all to gradeable events.
2029 is the year the ledger stops being a promise and becomes a report card.
The short answer, before the long one
Asked what Nostradamus predicted for 2029, the record answers in one word: nothing. Of the 942 quatrains he published, only a handful carry explicit dates — the seventh month of 1999 in quatrain X.72, October 1727 in quatrain III.77, and the year 3797 as the far horizon of the whole enterprise, stated in the preface to his son César. No verse names 2029. No verse names any year of the decade you are living through. Every headline claiming otherwise is a modern interpretation performed by a living author and published under a sixteenth-century byline.
This page therefore does something different from the genre it shares a search results page with. It is a computed reading: every forward-looking number below is a probabilistic simulation from this observatory's registry, sealed in July 2026 with an uncertainty range, an objective resolution test, and a written falsifiability condition. Sealed numbers never move — the annotations move around them, dated and sourced, so honesty has a timeline. None of it is prophecy, and none of it is financial, medical or life advice.
What makes 2029 worth the deepest dossier in this series is not any verse — it is arithmetic. Three of the ledger's largest seals resolve by the end of that year, the Congressional Budget Office's own baseline has federal debt crossing its 1946 record in that exact year, and several live projection windows slam shut in it. 2029 is not the year we predict. It is the year we are graded at scale — so this article shows, stage by stage, exactly how the 2029 readings were derived, in enough detail that you could recompute them against us.
2029: the year the ledger is judged at scale
On 3 July 2026 this observatory sealed a Calibration Ledger — locked probabilities, objective resolution criteria, hard resolve-by dates, all currently PENDING. Three of its heaviest entries resolve by the end of 2029. CAL-01 seals 78% that US federal debt held by the public exceeds 105% of GDP in an official FRED or CBO reading by 2029-12-31, from a baseline near 100%. CAL-02 seals 82% that through 2029-12-31 there is no exchange of live fire between US and PRC military forces — collisions, lasing and cyber operations excluded by written criteria; live fire only. CAL-03 seals 62% that a probability-sampled national survey finds more than 60% of US adults using generative-AI tools at least weekly by 2029-06-30, from a Pew baseline of 34% who had ever used ChatGPT in June 2025.
Do the multiplication the genre never does. If the three anchors were independent, the chance that all three base cases land — record debt, no live fire, majority-weekly AI — is 0.78 × 0.82 × 0.62, roughly 40%. A ledger built entirely of favorites still expects at least one public miss by December 2029. That is not a defect of the method; it is the method. A forecaster whose every call lands was never stating real probabilities.
Around the anchors, live projection windows close. TEC-03 (data-center power crunch, sealed 66%) and SOC-02 (age-gated social media as the OECD norm, 66%) both expire at end-2029, as do CRY-01 (ten nations holding strategic bitcoin reserves, 55%) and CRY-03 (a top-3 stablecoin breaking the buck, 30%). CAL-07's window — at least one of 2026 through 2029 setting a new global temperature record, sealed 76% — is fully spanned by the year. And by January 2029 the first two rounds of public Brier scoring, from July 2027 and 2028, will already be on the record: you will be able to read our grades before weighing a single number on this page.
The methodology: how a 2029 reading is computed
Every domain reading below is derived through the same four-stage chain, and each stage is published. Stage one is historical base rates: how often the event class actually occurred per decade across the nine eras indexed by our Historical Resonance Index, 1815 to 2000. Base rates set the prior — the number you would quote before looking at today's newspapers. Stage two is resonance: the index scores July 2026 against each era on five weighted components (debt load .20, great-power rivalry .25, internal polarization .20, tech-disruption velocity .20, monetary regime strain .15), and the eras that score closest tell you which base rates deserve extra weight.
Stage three is the Loom — the observatory's live state-space instrument: six threads, each with six named strands scored in [−1, +1], updated by the rule s′ = clamp(0.99·s + crossweave + wire impulses) with a hard per-step cap. Without fresh evidence a reading decays toward baseline on a roughly 17-day structural half-life — conviction fades slowly, never instantly, never never. The crossweave — 24 published couplings, each with a weight, a one-line rationale and a historical precedent — is deliberately bounded so that a sustained source at full strength can hold its target at no more than its own magnitude: influence, not amplification. Every transition is caused, logged, versioned and reversible.
Stage four is the sealed anchors: the CAL entries and live projections that resolve before or during 2029. These pin the interpretation to gradeable events and provide the update schedule — when CAL-13 (a 40-hour autonomous agent work-week by end-2027, sealed 75%) resolves, the 2029 labor reading hardens or cools accordingly; when CAL-12 (2026 among the two warmest years) is graded in early 2027, the 2029 climate reading inherits the verdict. One rule governs all four stages: no stage ever moves a sealed number. The seals are locked and will be scored as written. The 2029 reading is an interpretation layer computed on top of them, and wherever interpretation and seal diverge, we say so in the text.
Stage one and two: base rates, and the eras that get extra weight
The base rates, named. Systemic great-power war: of the nine indexed eras, two — 1910–14 and 1937–39 — were followed by systemic war within five years, a configuration rate near 22%; but both carried rivalry components of 88 and 92, and July 2026 reads 78. Monetary regime resets: three across the indexed span — the 1873 silver demonetization, the 1931–33 gold exits, the 1971 end of Bretton Woods — roughly one per five decades, with the hazard concentrating when debt and monetary-strain components both read above 70, as they do now (82 and 72). Reserve-currency debt above 100% of GDP: two episodes, Britain 1815–19 and the US 1946–51, and both exited through repression and inflation over decades — the default base rate at this position is zero for two. Carrier-technology installation busts: two of two — the 1873 railway bust and the 2000–01 fiber bust — produced a leverage crisis within two decades of installation onset.
Stage two says which of those rates to weight. The resonance table's top is crowded and split: 1968–74 scores 83, while 1910–14 and 1873–96 tie at 82. That triple is the whole 2029 problem in one row. The first resolved in reform and inflation, the second in continental war, the third in a two-decade deflation that coexisted with the century's fastest real growth. Nearly identical structural pressure, three different exits. Resonance identifies the fork, not the branch — and 2029 sits close enough to the fork that every domain reading below has to be priced across all three branches rather than betting the narrative on one.
Note what today's component readings emphasize. Tech-disruption velocity is the highest input on the board at 88 — above even 1873–96's 85 and 1999–2000's 85 — which is why the 1873 pattern (technology compounding while finance bleeds) earns extra weight in the money and technology readings below. Polarization at 84 sits nearest 1848's 90, the pamphlet-wave precedent. And the component 2026 has that 1873 lacked is the military axis: Britain faced no armed peer during its long deflation. That one gap is the standing argument against reading 2029 purely through the most economically comfortable analogue.
Stage three: the Loom's posture, and the couplings that decide 2029
As of the August 2026 seed state, the hottest strands on the entire instrument are machine capex-intensity at +0.75 (the AI build-out running at fiber-overbuild scale), social polarization at +0.70, machine capability-slope at +0.70 (METR-measured agent time-horizons growing near 10× a year), social trust-decay at +0.65, geopolitical arms-tempo at +0.65 (SIPRI's eleventh consecutive annual rise, $2.89T in 2025), and monetary debt-strain at +0.60. The most consequential negative readings are information integrity at −0.50, diplomacy momentum at −0.35, and institutional capacity at −0.30. Every value carries a one-line published rationale, reviewable like any other data on the platform.
The crossweave then says how these feed each other on the road to 2029, and each coupling names its precedent. Arms-tempo pressing on deterrence-stress (weight 0.04, precedent: the 1910–14 naval race) is the channel that keeps the Pacific reading warm even in quiet months. The one reliable damper — diplomacy bleeding escalation directly (−0.05, precedents: the 1962 back-channel and the 1987 INF treaty) — is currently running in reverse, because the diplomacy strand itself is negative. Debt-strain bending central banks toward the fisc (0.05, precedent: the 1942–51 Fed peg and the 1970s) is the quiet engine of the money reading. And capex-intensity leaking into credit-stress (0.03, precedent: 1873 rails and 2001 fiber) is the instrument's standing warning that the AI build-out is also a financial event.
Three more couplings carry the social and climate readings. Capability-slope converting to labor displacement with a lag (0.05, precedents: the looms of 1811 and the ATMs of the 1990s) connects the technology thread to the entry-level employment data. Labor displacement feeding generational strain (0.04, precedent: 1980s deindustrialization) is why the 2029 society reading is dominated by the young. And the climate chain — thermal anomaly loading the disaster dice (0.05, the attribution literature), disasters driving insurers out (0.05, Hurricane Andrew 1992 and Florida/California 2023), insurance retreat leaking into credit (0.03, the 2024 Senate climate-insurance file) — is the longest fully documented causal chain on the board, and every link of it is expected to be visibly load-bearing by 2029.
Geopolitics in 2029: the split matures, the taboo is graded
The structural reading first. GEO-01 seals 74% (±10) on the world economy splitting into rival trading blocs across 2026–2030, and by 2029 the projection expects the split to be adjudicable-in-progress rather than debatable: tariff walls (the US average effective rate near 18% in 2025, the highest since 1934 on the Yale Budget Lab's measure), export-control regimes and parallel settlement rails hardening into architecture, with India, Brazil and the Gulf deciding the depth of the cut. The base rate behind it is the bloc-formation record of the 1930s and the Cold War; the resonance weighting comes from 1910–14's demonstration that deep interdependence does not prevent bloc formation once security logic takes over.
The sharper edge is the Pacific, and here the ledger's construction deserves close reading. GEO-02 seals 38% (±15) on a direct US–China military incident producing casualties by 2031. CAL-02 seals 82% that no live fire is exchanged through 2029. These are consistent, not contradictory: GEO-02's definition includes collisions and shoot-downs across a six-year window, while CAL-02 counts live fire only and closes two years earlier. Spread GEO-02's sealed mass across its window against the current escalation strand (+0.55) and the derived 2029-alone incident hazard lands in the high single digits — derived, not sealed. The Loom explains why the reading stays warm: arms-tempo at +0.65 presses deterrence-stress through the naval-race coupling, while the diplomacy damper is running in reverse at −0.35. July 2026's Taiwan Strait live-fire drills, staged the day after high-level talks, are exactly what that configuration looks like in the wire.
What would falsify the domain reading. GEO-01 is wrong if the US and China execute a broad tariff rollback and world goods-trade growth runs above 4% a year for two consecutive years before 2029. CAL-02 scores INCORRECT the day any confirmed US–PRC live-fire exchange occurs — one event, objectively adjudicated by governments or three major news organizations. And the entire escalation reading cools if PLA air-incursion tempo falls back below its 2021 baseline for a sustained year. Each line was written before the outcome, which is the only time such lines are worth writing.
Money and debt in 2029: the year the record breaks
2029 is the single most legible year in the entire fiscal projection set, because the Congressional Budget Office put it on the calendar itself: in the CBO's long-term baseline, federal debt held by the public crosses its 1946 record — roughly 106% of GDP — in 2029, reaching about 107%. CAL-01 seals 78% that an official reading prints at or above 105% by 2029-12-31, from a baseline near 100% (the Q1-2026 FRED reading was 98.7%). The residual 22% is not decoration: it prices a recession-flattened denominator, a genuine consolidation, or simple measurement timing. Around the threshold sit the numbers that make it bind — net interest near $1 trillion a year, more than defense; Moody's removal of the last AAA in May 2025; and 10Y/30Y yields near 20-year highs as of August 2026, with the Treasury doubling long-end buybacks.
The stress reading rides on top of the level reading. FIN-01 seals 62% (±12) that a US sovereign-debt stress event forces a fiscal regime change by 2030 — and 2029 is where the derived hazard concentrates, because it is the year the record breaks in public while the 2031–2034 debt-service inflection window, where rolling maturities and interest costs converge on the discretionary budget, moves from projection to planning horizon. The base rate from stage one is blunt: both prior reserve-currency episodes at this debt position, 1815–19 and 1946–51, exited through financial repression and inflation, not default — which is why the Loom's debt-strain strand (+0.60) feeding fiscal dominance (+0.40) through the 1942–51 Fed-peg coupling is the single most watched channel on the monetary thread. The dollar reading stays deliberately two-sided: FIN-03 seals 45% on the reserve share breaking 50% by 2032, while CAL-06 seals only 8% on a fall below 45% by 2030 — erosion, not collapse, and the Q1-2026 COFER share actually ticked up to 57.13%.
What would falsify it. FIN-01's pre-registered refutation: deficits under 4% of GDP for two consecutive years through 2030, compressed term premia, and no Treasury auction stress episode. CAL-01 simply fails if no quarterly reading reaches 105% by end-2029 — arithmetic, no interpretation. And the repression thesis is discounted if central-bank gold demand — 1,000+ tonnes a year for three straight years before cooling to 345t in H1 2026 — keeps fading while real yields stay positive. The sealed numbers will not move on any of this; they will be graded against it.
AI and technology in 2029: diffusion gets its verdict
The 2029 technology reading is a diffusion reading, not a capability reading, and it is anchored by CAL-03: 62% that more than 60% of US adults report weekly generative-AI use by mid-2029, in a probability-sampled national survey. The baseline arc is already the steepest in the adoption record — from Pew's 34% ever-used figure in June 2025 to the St. Louis Fed's finding that gen-AI reached a majority of US adults 18–64 in under three years, the fastest adoption curve ever recorded. The seal at 62% rather than 90% prices the last mile honestly: weekly use among all adults, including the retired and the offline, is a far higher bar than ever-use among the working-age.
Above the anchor sits the labor projection. TEC-01 seals 70% (±12) on AI automating more than a quarter of white-collar task-hours by 2030, with the entry rung breaking first — the Stanford Digital Economy Lab's ~13% relative decline in early-career employment in AI-exposed occupations is the leading indicator already in print, and the Loom's capability-slope strand (+0.70) feeds labor displacement through the oldest coupling on the board (looms 1811, ATMs 1990s). The update schedule runs through CAL-13: if the 40-hour autonomous agent work-week resolves CORRECT by end-2027 — as of August 2026 METR's operative data shows ~10×-per-year time-horizon growth with top models at 16–20 hours — the 2029 reading hardens; if it resolves INCORRECT, TEC-01's mid-window evidence weakens and we will say so in the field notes, with the sealed 70% standing for the grade.
The physical bill comes due in 2029 by construction: TEC-03's window (66%, ±10) closes at end-2029. The published trajectory is Lawrence Berkeley National Laboratory's: data centers drew 4.4% of US electricity in 2023, projected to 6.7–12% by 2028, against transformers and transmission that move at permitting speed. And the crossweave adds the financial warning the headlines omit: capex-intensity is the hottest strand on the entire instrument at +0.75, and its coupling into credit-stress carries the two precedents — 1873 rails, 2001 fiber — where the carrier technology survived while the leverage behind it did not. Falsification: TEC-03 fails if 2029 closes without a documented grid-constraint event forcing curtailment or major buildout deferral; TEC-01's refutation is early-career employment recovering to its 2022 trend with measured automation under 15%; CAL-03 fails on the survey arithmetic alone.
Climate in 2029: the record-year window closes
The cleanest climate seal spans exactly the years this dossier covers: CAL-07 seals 76% that at least one of 2026 through 2029 sets a new global temperature record, against a 2024 record holder near 1.55°C above pre-industrial — the first calendar year above 1.5°C. The derived per-year record hazard runs near 30% and is not uniform: it loads onto El Niño-boosted years, and the ENSO strand reads +0.35 with an event developing as of mid-2026. The nearest update arrives fast — CAL-12 (2026 among the two warmest years, sealed 66%) is currently AT RISK with Jan–Jul 2026 running third, and if it scores INCORRECT in early 2027, as we may well publicly do, the Bayesian weight inside CAL-07 shifts onto 2027–2028 rather than diminishing the window seal, which stands as written.
The economic transmission is the part the prophecy genre never prices. CLI-01 seals 62% (±10) on insurance retreat triggering a climate repricing of housing in a G7 economy by 2030 — actuarial, not meteorological — and the Loom carries it as the longest documented chain on the board: thermal anomaly (+0.55) loads disaster frequency (+0.50), disasters drive underwriters out (insurance-retreat +0.60, with non-renewals already tripled in top-risk counties), and uninsurable collateral leaks into credit through the coupling whose precedent is the 2024 Senate climate-insurance file. By 2029 the reading expects this chain to be visible in mortgage pricing, not just in premiums. The upside is sealed with the same discipline: CAL-10 gives 66% to a 900 GW solar year by 2030, from a 599 GW baseline in 2024.
Falsification, per line. CAL-07 fails if neither NASA GISTEMP nor NOAA declares any of 2026–2029 a record year. CLI-01's refutation is stabilizing insurance availability without state backstops absorbing the risk. GEO-04 — the food-system spillover, sealed 48% on acute insecurity passing 400 million within 2026–2030, from the 295 million recorded across 53 countries in 2024 — is wrong if the Global Report on Food Crises records two consecutive declining years before 2030 with no simultaneous two-breadbasket failure. The sixth consecutive annual rise says the burden of proof currently sits with the optimists.
Society and demography in 2029: the young carry the reading
The social thread enters 2029 with the most extreme posture on the instrument: polarization +0.70 (the 84/100 resonance component, nearest 1848's pamphlet-wave 90), trust-decay +0.65 (federal trust in the 17–22% band, near historic lows on Pew's series), and information integrity at −0.50 and degrading. The regulatory response is the domain's clearest 2029 event: SOC-02 seals 66% that age-gated social media becomes the OECD norm by end-2029 — the tobacco moment, running on the template Australia set in motion — and its window closes in the year itself, so 2029 grades it either way.
The deeper 2029 reading is generational, and it is computed from a coupling rather than a headline. Labor displacement (+0.40, with the entry-rung erosion already in the Stanford data) feeds generational strain (+0.50) through the channel whose precedent is 1980s deindustrialization — the last time a cohort's economic ladder was removed faster than institutions could rebuild it. SOC-03 seals 60% that the generational wealth rupture becomes an organized political movement by 2032; by 2029, mid-window, the projection expects organized form — candidacies, tenant and debtor blocs, platform planks — not merely sentiment. Beneath it runs the demographic floor: China's 7.9 million births in the latest year (−17% year on year) and the global trough thesis SOC-04 tracks, slow variables that set the 2030s' labor and fiscal arithmetic regardless of any single election.
Falsification. SOC-02 fails if fewer than a plurality of OECD members have binding age-gating in force by end-2029. SOC-03's refutation is early-career employment and first-buyer age both recovering toward their pre-2022 trends without a transfer-program response — the world where the ladder repairs itself. And the polarization reading cools if the trust series posts a sustained multi-year recovery, something it has not done in two decades. We seal no prediction on the 2028 US election or its 2029 aftermath — no winner, no party, no legitimacy call — for the reason stated in the 2028 dossier: no objective resolution test can honestly be written years before nominees exist, and no number will be retrofitted later.
What we will not pretend to know about 2029
First, the K-wave positioning question. Whether the late 2020s are the winter of one Kondratieff long wave or the installation spring of the next is the single largest open structural question in the framework stack, and the engine declines to date the turn. The honest evidence cuts both ways: 1873–96 shows deflationary finance and the century's fastest real growth coexisting for twenty years, and the AI build-out at +0.75 capex intensity is equally consistent with a Perez installation mania before a bust and with a productivity surge already underway. Any site that tells you which with confidence is selling narrative, not measurement.
Second, the Moon. CAL-09 sealed 25% on a crewed NASA lunar landing by end-2028 — deliberately against the official schedule — and the 2026 replan that moved the crewed landing to roughly 2028 has already vindicated the skepticism premium. That makes 2029 the slip-catcher year: the window in which the landing either finally happens or the decade's schedule collapses outright. We sealed no 2029 lunar entry, so we publish no number for it — the rarest sentence in the prediction business remains 'we have no number for that.' Third, the nuclear line: GEO-03's 8% on the taboo breaking is a 2026–2035 reading, not a 2029 call, and slicing single years off a decade-scale ruin risk would be false precision in the one domain where false precision is most dangerous.
What remains is the appointment. By January 2029, two full rounds of public Brier scoring — July 2027 and 2028 — will already be on the record, misses printed with the same prominence as hits. By 31 December 2029, CAL-01, CAL-02 and CAL-03 resolve, the largest single scoring event the ledger will have faced: the debt record, the Pacific taboo, the diffusion threshold, graded in arithmetic. Nostradamus has had 471 years without one scored prediction. This page will be scored whether we like the results or not — and that difference, not any verse, is the entire proposition. Nothing here is advice; all of it is checkable.
Frequently asked
What did Nostradamus predict for 2029?
Nothing. The 942 quatrains contain almost no explicit dates — the seventh month of 1999 (X.72), October 1727 (III.77), and a horizon of 3797 stated in the preface to his son César. No text Nostradamus wrote names 2029 or any event of the year. Every claim otherwise is a modern interpretation retrofitted onto undated sixteenth-century verse.
What is most likely to happen in 2029?
According to this observatory's sealed probabilistic simulations — not prophecies — the leading readings whose windows cover 2029 are: no US–China live-fire exchange (82%), US debt held by the public exceeding 105% of GDP (78%), a new global temperature record within 2026–2029 (76%), the trading-bloc split maturing (74%), AI automating over a quarter of white-collar task-hours by 2030 (70%), and a data-center grid crunch by end-2029 (66%). Each was sealed in July 2026 with an uncertainty range and a written falsifiability condition, and some will be wrong.
Will US debt hit a record in 2029?
The CBO's own baseline has federal debt held by the public crossing its 1946 record (~106% of GDP) in 2029, reaching about 107%. This platform sealed 78% (CAL-01) that an official reading prints at or above 105% by 2029-12-31, from a baseline near 100%. The remaining 22% prices a recession-flattened denominator or genuine consolidation. It is a probabilistic simulation, graded by arithmetic at year-end 2029 — not a certainty and not financial advice.
Will there be a war between the US and China by 2029?
The sealed base case says no live fire: CAL-02 assigns 82% to no exchange of live fire between US and PRC forces through 2029, under written criteria (collisions, lasing and cyber operations do not count). Separately, GEO-02 assigns 38% to a broader military incident producing casualties across 2026–2031. The two are consistent — different definitions, different windows — and both were sealed in July 2026 with falsifiability lines.
Will most people be using AI by 2029?
CAL-03 seals 62% that more than 60% of US adults report weekly generative-AI use in a probability-sampled national survey by mid-2029, from a Pew baseline of 34% ever-use in June 2025. Adoption has run at record speed — a majority of adults 18–64 within three years per the St. Louis Fed — but weekly use across all adults is a materially higher bar, which is why the seal is 62% and not 90%.
How can I check whether these 2029 predictions come true?
Through the public Calibration Ledger, sealed 2026-07-03 and 2026-07-11: sixteen entries with locked probabilities, objective resolution criteria and hard resolve-by dates, all currently PENDING. First Brier scoring publishes in July 2027, with misses given equal prominence to hits — so by the time 2029 arrives, two rounds of grades will already be public. CAL-01, CAL-02 and CAL-03 all resolve by end-2029; read the grades before trusting any number on this page.