Speculations

What is dark energy?

The quintom branch of bounce cosmology is a theoretical possibility that predicts w-crossing. External DESI analyses are not a BigBounce result, and our program has not run a free-w0-wa analysis. P1B is namaster-proof research software, not an MCMC paper. The actionable next step would be a separately scoped in-house analysis.

f_NL = -35/16: a conditional future test

P2 derives a conditional matter-contraction prediction f_NL = -2.1875. SPHEREx may provide a future test, but no present measurement proves a bounce or resolves the model landscape by itself.

What is dark matter?

Some bounce scenarios motivate asteroid-mass primordial black holes as dark-matter candidates. This remains a theoretical research direction, not a conclusion from the program's surveys or current artifacts.

Can an accelerating universe recollapse?

A constant positive cosmological constant at the observed matter density (Ω_m ≈ 0.315) forbids a turnaround at any curvature — not merely improbable, mathematically impossible. Only a dark-energy density that crosses zero in the future — zero-crossing quintessence, an interacting dark sector, or effective-SEC-restoring modified gravity — could permit one. DESI DR2's Lyα full-shape result (arXiv:2607.27410) neither implies nor excludes such models; a CPL (w₀, w_a) fit extrapolated to a→∞ is a fitting-function artifact, not a fate forecast.

Crunch is not bounce: which structure does 'born in a black hole' mean?

A future contraction does not itself produce a new universe — classical GR ends contraction in a singularity, the opposite sign from what a bounce requires. Of five constructions the 2025 notes conflated, only a closed FLRW region behind the horizon (Popławski; Frolov–Markov–Mukhanov) matches the phrase 'born in a black hole.' That daughter is causally sealed: no exterior observable exists. The spin-axis inheritance route some of the same notes hoped for was excluded against DESI data on 2026-09-02.

Does a daughter universe hold more matter than its parent black hole?

An independent reproduction of Popławski's Einstein–Cartan daughter-universe model (ApJ 832:96, 2016): 15 of 26 published steps reproduce exactly, 8 reproduce only with an unstated assumption, and 3 do not reproduce. The daughter can hold arbitrarily more proper matter than the parent's mass — but the excess is set by a free particle-production coefficient, not by the parent. No invariant parent-to-child amplification ratio exists. Local conservation, ∇_μT^{μν}=0, is respected throughout.

Does bounce-compatible dark energy leave a fingerprint we can already see?

A future turnaround — the universe stopping its expansion and contracting — is impossible under a constant, positive cosmological constant, so it requires the dark-energy density to fall to zero and go negative. We crossed the dark-energy classes that are compatible with a bounce against that requirement and found three that qualify: quintessence and k-essence whose potential crosses zero, and a dark sector that steadily drains energy into dark matter. We then asked whether today's data can tell any of them apart from ΛCDM. They can — but only above a threshold, and the threshold is in the present-day equation of state, not in the fate. At DESI DR2 precision the separation reaches 1σ once w₀ ≳ −0.95, while the turnaround epoch that same threshold corresponds to shifts by a factor of four (54 Gyr vs 14 Gyr) depending only on the shape of the potential assumed. Current data constrain how fast dark energy is evolving now; they do not measure when, or whether, the expansion ends.

GW echoes from black hole mergers

If black hole interiors contain baby universes, mergers might produce gravitational wave echoes. LIGO/Virgo/KAGRA could detect these in existing data.

Supermassive black hole formation

Early massive black holes motivate several formation hypotheses, including speculative primordial seeds. The legacy anomaly archive is not a discovery catalog and should not be treated as evidence for any such mechanism.

Exploratory anomaly candidates

Historic exploratory autoencoder runs flagged archival candidates for possible follow-up. They are not confirmed detections and are now legacy/superseded pipeline records. P3 is the integrated Supporting Data Release recovering 181 DESI DR1 TARGETIDs from a frozen historical list, not a standalone discovery claim.

Galaxy morphology: what shapes don't fit?

An autoencoder on galaxy images could find unusual morphologies: double nuclei, tidal tails, ring galaxies, or entirely novel structures.

Time-domain: what changed?

A future cross-match could identify candidates that are unusual in more than one modality. Such candidates would still require independent validation and would not by themselves support a bounce claim.

Dyson spheres at scale

A Dyson sphere would create anomalous IR excess. Prior searches found 7 candidates from 5M stars. We could search 100x more with our pipeline.

FRB anomaly detection

Fast radio bursts from cosmological distances — our anomaly detection approach on FRB properties could identify subpopulations that current classification misses.

LHC anomaly detection

An unsupervised autoencoder on collision events could find signatures that don't match any Standard Model prediction — new particles hiding in plain sight.

Spacetime fabric: discrete or continuous?

LQG predicts discrete spacetime producing energy-dependent photon speeds. Our birefringence prediction is another test of spacetime microstructure.

Multi-messenger astronomy

Combining heterogeneous data could prioritize future follow-up candidates. Agreement across pipelines is not a discovery confirmation or evidence for a bounce without independent validation.

The Hubify lab scaling vision

Build reproducible discovery infrastructure around explicit parent-catalog, model, score, and selection lineage. Any future outputs should be scoped as question-first programs rather than a paper-count target.