Civilisation continuity requires preserving both species and genetic variance, not just specimens. The current conservation floor is bleak: the IUCN Red List now covers 175,909 species, with 49,505 threatened with extinction; in 2025 IUCN reported more than 47,000 threatened species, and 2026 updates pushed the total higher[1][2].
1) How many species are at risk
The best hard number for the current global extinction-risk burden is 49,505 threatened species on the IUCN Red List as of the 2026 update[1][6]. That figure is the operational minimum for “known species at extinction risk,” but it understates reality because many taxa remain unevaluated or data-deficient[2].
Additional high-risk signals from 2026 illustrate the accelerating loss:
- 62% of endemic hydrothermal vent molluscs assessed worldwide are at risk from deep-sea mining, 125 of 201 species[1].
- A 2026 assessment found 20% of assessed soil species at risk, with another 20% data deficient[3].
- European freshwater fishes now show 42% threatened and 18% near threatened among 558 native species[5].
2) Genome banking vs seed banking
A seed bank stores seeds of plants, usually dried and kept cold, to preserve plant species and crop varieties for later propagation. A genome bank is broader: it stores DNA, tissue, gametes, embryos, viable cells, or other genetic material from animals, plants, fungi, and microbes for future sequencing, cloning, assisted reproduction, or synthetic reconstruction.
Operational difference:
- Seed bank = best for orthodox seeds that tolerate drying and freezing.
- Genome bank = essential for animals and many plants that cannot be preserved as seed, plus all species whose reproductive biology makes seed banking impossible.
For civilisation backup, seed banks preserve ecosystem restoration capacity for plants; genome banks preserve recoverability of genetic lineages across the tree of life.
3) The Frozen Ark project
The Frozen Ark Project is a global collaboration to safeguard the DNA, tissue, and viable cells of endangered animals. Its core purpose is to bank genetic material before extinction erases it. The project explicitly focuses on deep-freezing endangered species’ genetic materials through biobanking and coordination.
Strategic value:
- Captures genetic material from animals that cannot be preserved by seed banking.
- Provides a reference archive for future genomics, reproductive technology, and potential de-extinction work.
- Works as a “last-record” archive even when whole organisms are lost.
4) DNA storage media longevity
### Silicon chips
Silicon-based DNA storage is attractive for density and easy electronic retrieval, but it is not yet the long-duration, field-proven archival medium of a frozen biobank. Its main advantage is high information density and fast access; its weakness is that long-term reliability depends on device integrity, power, and error correction. For civilisation backup, silicon is best treated as an index and access layer, not the primary biological archive.
### Synthetic DNA
Synthetic DNA storage is the most biologically direct archival form for sequence information. DNA can, in principle, store enormous information density, but survival depends on chemical stability, dryness, cold, oxygen exclusion, and error correction. The practical advantage is that DNA is the same molecule life uses; the practical weakness is that retrieval and read/write costs remain high and physical degradation is still a concern.
### Crystal storage
Crystal or glass-like storage media are designed for extreme longevity because information is embedded in highly stable solid structures. For long-term civilisational continuity, crystal storage is the best bet for metadata, genome indexes, and reference sequence libraries, especially where power or maintenance may fail. The limitation is that it stores information about genomes, not the biological material itself.
Bottom line:
- Best for actual recoverable biology: frozen cells, tissues, gametes, embryos.
- Best for sequence archives: synthetic DNA.
- Best for ultra-long metadata durability: crystal storage.
- Best for active curation and rapid retrieval: silicon systems.
A serious planetary backup should use all four, with layered redundancy.
5) De-extinction feasibility
De-extinction is technically plausible in limited cases but not a substitute for conservation. It is most feasible when:
- A close living relative exists.
- High-quality genome data exist.
- At least some preserved cells, nuclei, or viable tissue survive.
- A reproductive pathway can be engineered or borrowed.
Most realistic near-term outcomes are:
- Back-breeding of lost traits.
- Gene editing in close relatives.
- Cloning only where intact cells/nuclei survive.
Limits:
- It does not restore the original ecosystem.
- It rarely recreates the original genome with perfect fidelity.
- It cannot recover lost epigenetics, microbiomes, learned behavior, or ecological context.
- It risks diverting resources from preventing extinctions that are still avoidable.
Civilisation rule: de-extinction is a repair tool, not a preservation strategy.
6) Minimum viable genetic diversity
The minimum viable population is not just about headcount; it is about heterozygosity, effective population size, and inbreeding control. A population can be numerically present and still be genetically dead.
For long-term survivability, the archive should aim for:
- Multiple unrelated founders, not a single pair.
- Enough representation to capture rare alleles across subpopulations.
- Preservation of sex cell diversity and population structure.
- Repeated sampling over time, because one snapshot misses evolving variation.
Useful operational thresholds:
- For short-term avoidance of immediate inbreeding collapse, wildlife management often targets an effective population size (Ne) of at least 50–100.
- For longer-term adaptive potential, targets are commonly Ne in the hundreds to thousands.
- For civilisational continuity, the archive should store genetic material from **dozens to hundreds of