Python

Gojo's Infinity, Explained With Real Math

Gojo Satoru's “Infinity” examined through four independent mathematical lenses — can an attacker ever reach him?

gojo_infinity emblem — geometric abstraction of converging geodesics
infinity-lab — gojo_infinity emblem

The idea

In Jujutsu Kaisen, Gojo Satoru maintains a perpetual barrier called “Infinity” that subdivides the space between any incoming attack and himself into infinitely many sub-intervals, so the attacker — in theory — never arrives. The show treats this as unbreakable.

This package asks: is that actually true? The answer depends entirely on which mathematics you apply. Four independent lenses each give a different verdict, and the verdicts disagree. The project is inspired by Achmad Roykhan Sabiq’s 2026 Oxford Mathematics Essay Competition entry and the 2021 RIKEN × Gege Akutami “Abyss of Math” paper.

All arithmetic uses Python’s fractions.Fraction throughout — no floating point in any comparison or convergence test. The package ships 155 pure-stdlib tests and 217 tests with numpy/matplotlib.

What is a convergent geometric series?

The intuition behind “Infinity” is a version of Zeno’s paradox: to reach Gojo you must first cross half the gap, then half of what remains, then half again — an endless sequence of ever-shorter steps. Zeno concluded that motion is therefore impossible. Real analysis disagrees, and the reason is convergence.

A geometric series adds terms that each shrink by a constant ratio r. When |r| < 1, the infinite sum has a finite value: ∑ rⁿ = 1/(1−r). For the half-the-remaining-distance model r = 1/2, so 1/2 + 1/4 + 1/8 + … = 1. The partial sums Sₙ = 1−(1/2)ⁿ climb toward exactly 1 and never overshoot. So even though there are infinitely many steps, their total length — and, at constant speed, the total time — is finite. “Never arriving” is the wrong conclusion: the attacker arrives at the sum of the series.

That is why two of the four lenses rate Infinity Fragile. To make “never arriving” true you cannot merely subdivide distance — you must change the geometry itself. The Riemannian lens does exactly that: it replaces the flat ruler with a conformal metric Ω(x) that diverges as x approaches Gojo’s position. Now the length of the path is ∫ Ω(x) dx, and this integral goes to infinity before contact. A divergent length integral, unlike a convergent geometric series, genuinely means no finite-length geodesic ever reaches him. Convergence lets the attacker in; divergence keeps him out. Which one applies depends entirely on the physical model you choose.

Gojo's Infinity questions

Can Gojo's Infinity actually stop an attacker, mathematically?

It depends on the model. Under a convergent geometric (Zeno) series the partial sums 1−(1/2)ⁿ reach 1, so the attacker arrives in finite time. Only a conformal Riemannian metric that blows up near Gojo makes the geodesic length infinite and genuinely prevents arrival.

Why can infinitely many steps still finish in finite time?

Because the step sizes shrink geometrically. The sum 1/2 + 1/4 + 1/8 + … converges to 1, so infinitely many terms add to a finite total. Covering ever-smaller distances does not prevent reaching the target.

What is the geodesic metric-blowup view of Infinity?

Instead of subdividing distance, you change the ruler. A conformal factor that diverges as the attacker approaches Gojo’s position makes the length integral of the metric infinite, so no finite-length path reaches him. This is the one lens where Infinity holds.

The mathematics

Lens 1 — Zeno’s geometric series — Fragile

Model each step as covering half the remaining distance. The partial sums converge:

Geometric series
S_n = 1 - (1/2)^n  →  1  as  n → ∞

The series sums to exactly 1.  Under real analysis the attacker
arrives in finite time — the infinite subdivision does not stop them.

Lens 2 — Lebesgue measure — Fragile

Build the Infinity barrier as a Cantor-like construction: iteratively remove the middle third of each remaining interval. The resulting set Z has measure zero.

Lebesgue measure
m(Z) = 0

A measure-zero set has zero thickness — a point mass moving at any
positive speed crosses it in zero time.  The barrier is invisible
to Lebesgue integration.

Lens 3 — Riemannian conformal geometry — Formidable

Replace the flat metric with a conformal metric that blows up as the attacker approaches Gojo’s position x₀. The conformal factor Ω is calibrated by bisection so that the metric at 80% approach equals 4.1 (a value extracted from the show’s geometry).

Conformal metric
Ω(x) = 1 + λ · K(x, x_gojo) / (x_gojo - x)

where λ ≈ 0.284118  (calibrated by bisection to g(0.8) = 4.1)

Geodesic length integral:
  L = ∫₀^{x_gojo} Ω(x) dx  →  math.inf

The path length diverges before contact.  Under this metric,
the attacker never arrives — the geometry itself is the barrier.

Lens 4 — Topology / World-Cutting Slash — Falls

Gojo’s “World-Cutting Slash” severs the conformal metric entirely, setting the metric to None at his position. The geodesic computation raises an exception and returns no path.

Severed metric
g(x_gojo) = None   (metric excised)

The topological surgery removes the singularity itself.
The barrier collapses — the Slash lands.

Verdict table

Lens Mathematical tool Key result Verdict
1 — Zeno Geometric series Sₙ → 1; attacker arrives Fragile
2 — Measure Lebesgue measure m(Z) = 0; barrier is null Fragile
3 — Geometry Conformal Riemannian metric L = ∞; path diverges Formidable
4 — Topology Metric excision (slash) g = None; barrier severed Falls

How it works

The package is a pure Python src-layout library with four independent modules, each implementing one lens. A top-level runner executes all four in sequence and prints both a verdict table and calls matplotlib to render the artifact plots.

  • Zeno module: iterates Fraction(1,2)**n steps, computes partial sums exactly, and checks convergence against Fraction(1,1).
  • Measure module: builds the Cantor set by iterating complement intervals, computes total measure as a Fraction, and confirms it equals zero.
  • Conformal geometry module: integrates the metric numerically using fractions.Fraction-based midpoint quadrature; the integral diverges (raises OverflowError or returns math.inf) before x₀.
  • Topology module: applies the World-Cutting Slash by setting metric[x_gojo] = None and re-integrating; the geodesic computation raises MetricExcisedError.

A shared calibrate.py bisects over λ to find the value that makes the conformal metric read 4.1 at 80% approach, matching the show’s implied geometry. All calibration comparisons use fractions.Fraction to avoid floating-point drift.

Figures & animations

Nine rendered artifacts: six static plots covering each lens, plus three animations showing the conformal geometry in motion. The four-lenses composite is the headline render.

Four lenses — composite animation

All four mathematical analyses side-by-side: series convergence, measure, blowing-up metric, and severed geodesic.

Geodesic approach animation

A geodesic advancing toward x₀ under the conformal metric — watch the step sizes shrink as the metric diverges.

Geodesic bundle, rotating (3-D)

The bundle of geodesics approaching the conformal singularity, rendered as a rotating 3-D surface. The length integral diverges before the apex.

Conformal metric Ω(x) blowing up as x approaches x_gojo

Metric blow-up

The conformal factor Ω(x) as the attacker approaches x₀. The metric diverges — the path length becomes infinite.

Zeno partial sums S_n converging to 1

Zeno series convergence

Partial sums Sₙ = 1 − (1/2)ⁿ approaching 1. Convergence is exact — the attacker reaches Gojo under real analysis.

Bundle of geodesics approaching the metric singularity

Geodesic bundle (2-D)

Multiple geodesic paths launched from different angles, all failing to reach x₀ in finite length.

Accumulated path length diverging to infinity

Length divergence

Accumulated geodesic length as a function of how far the attacker has travelled. The curve goes vertical before x₀.

3-D view of the conformal metric surface

Conformal surface (3-D static)

The metric surface Ω(x, y) rendered as a 3-D plot, showing the spike at Gojo’s position.

Cover convergence for the measure-zero Cantor construction

Cover convergence (Lebesgue)

Successive open covers of the Infinity barrier set, shrinking toward total measure zero.

Honest caveats: The Zeno and Lebesgue lenses both find Infinity fragile under standard real analysis. Only the Riemannian conformal model genuinely prevents arrival, and that model is a deliberate construction — the λ parameter is calibrated to match the show’s implied scale, not derived from first principles. The topology lens shows that even the conformal barrier is defeated by the World-Cutting Slash: severing the metric removes the singularity itself. Two of four lenses say “Fragile”; one says “Formidable”; one says “Falls.” The math cannot give a single answer because the question is underdetermined without fixing the physical model.

Run it

# Clone and enter the monorepo
git clone https://github.com/caelum0x/awesome-mad-projects.git
cd awesome-mad-projects/infinity-lab

# Install gojo_infinity in editable mode (no extra deps required)
pip install -e packages/gojo_infinity

# Run all four lenses and print verdict table
python -m gojo_infinity

# Run the full test suite (155 stdlib tests)
python -m pytest packages/gojo_infinity/tests -v

# Run tests including numpy/matplotlib artifacts (217 tests)
pip install numpy matplotlib
python -m pytest packages/gojo_infinity/tests -v