---
title: "Xenoplex"
description: "Connor Cook and Darren Zhu propose the Xenobiology Multiplex, or Xenoplex, as a research complex for investigating how life begins and continues to generate novelty."
type: "reading-notes"
authors:
  - "Connor Cook"
  - "Darren Zhu"
published: "2023"
original_url: "https://xenoplex.org/"
canonical_url: "https://antikythera.wiki/work/studio/xenoplex"
md_url: "https://antikythera.wiki/md/work/studio/xenoplex"
last_updated: "2026-09-06"
site: "Antikythera Wiki"
---

# Xenoplex

> Independent, unofficial reading notes from Antikythera Wiki (https://antikythera.wiki). Written with AI from the published text, with citations; not by the work's author and not affiliated with Antikythera (https://antikythera.org). Read the original at the link below.

- **Authors:** Connor Cook, Darren Zhu
- **Published:** 2023
- **Kind:** Studio projects · Studio paper
- **Venue:** Planetary Computation Studio
- **Original:** https://xenoplex.org/
- **Length:** 2.5k words
- **This page:** https://antikythera.wiki/work/studio/xenoplex
- **Notes generated:** 2026-09-06

## Summary

[Connor Cook](https://antikythera.wiki/people/connor-cook) and [Darren Zhu](https://antikythera.wiki/people/darren-zhu) propose the Xenobiology Multiplex, or Xenoplex, as a research complex for investigating how life begins and continues to generate novelty. An array of chemical reactors would test conditions associated with early Earth and possible extraterrestrial environments, while algorithms would measure the results and adjust subsequent experiments. The project treats planetary computation as a recent evolutionary development that might investigate evolution's earliest transition. Its five possible outcomes range from experimental failure to the emergence of life in the coordinating algorithms. Xenoplex is a speculative experimental design: the essay reports no constructed array, experimental dataset, or demonstrated origin of life.

## The argument

Cook and Zhu begin by placing computers within a longer history of information processing. Genes, proteins, and cells process information before electronic computing exists. Drawing on John Maynard Smith and Eörs Szathmáry, they describe evolution through transitions that change both the unit of individuality and the means of storing and transmitting information. Multicellularity reorganizes the individual; genetic inheritance and natural language reorganize informational capacity. The transition from natural to machinic language may mark a further change in what evolution can produce.

The proposed experiment turns that continuity into a research method. Planetary computation would investigate its own conditions of emergence by reconstructing possible beginnings of life. Comparing many trajectories could distinguish features peculiar to terrestrial history from more general evolutionary mechanisms. Exoplanet conditions matter because replaying only Earth's history cannot establish how necessary that history was. The authors describe an evolutionary time-compression device, but provide no demonstrated acceleration rate or experimental timescale.

The array would combine physical chemistry with computational search. Differences in temperature and chemical concentration provide free energy that emerging structures might dissipate through self-organizing processes. Algorithms would vary conditions across reactors, while measurements from actual reactions would guide subsequent choices. Physical experiments are necessary because a software chemistry already reduces the possibilities of the material it represents. The proposal therefore depends on chemistry contributing behaviors that the initial model does not contain.

Recognition becomes as difficult as generation. A system optimized for a fixed target might produce only the forms of life its designers already expect. Cook and Zhu propose several changing measures of lifelikeness: chemical complexity, copy number, and molecular diversity. Conditions associated with promising behavior would propagate between reactors. They acknowledge that no exact measure of open-ended evolution exists, leaving unresolved how their optimization procedures would produce novelty beyond the chosen criteria.

The five scenarios preserve this uncertainty. Failure might expose the inadequacy of laboratory isolation; apparently terrestrial results might expose inadequate detectors. Alien reactions might produce unfamiliar living structures, while the most consequential process might emerge across the computational system coordinating them. These are interpretations of hypothetical outcomes, not findings. The project's strongest common claim is that producing and recognizing unfamiliar life could change what researchers mean by life in the first place.

## Section by section

### Introduction

The opening uses climate simulation as an example of planetary computation making planetary change perceptible. Cook and Zhu extend this reflexive relation backward: a computational system emerging through evolution could study the processes that made its emergence possible. The theory of major transitions supplies a connection between origins-of-life chemistry and contemporary computing. Open-ended evolution names the sustained production of adaptive novelty or complexity, rather than progress toward a predetermined endpoint.

### Experimental Design

Each reactor would begin with a different combination of chemicals and temperatures, drawing on reconstructed prebiotic conditions and putative exoplanet biosignatures. The authors invoke telescopes such as James Webb as sources for these constraints; this is a proposed use of observational information, not evidence that extraterrestrial life has been detected. Parallel gradient descent, a method of iteratively adjusting parameters to improve an objective, would search for productive environmental configurations.

The physical container itself is an experimental variable. Cook and Zhu cite research on borosilicate glass catalyzing reactions in the Miller–Urey apparatus to show how equipment can affect apparent origin-of-life processes. Each box would also connect to a mass spectrometer for sampling its chemical contents. The molecular assembly index associated with Sara Walker and [Lee Cronin](https://antikythera.wiki/people/leroy-cronin) is one proposed complexity measure, supplemented by copy number and diversity. Feedback would adjust nested goals and transfer successful conditions between neighboring reactors. The essay sketches this architecture without specifying a complete measurement or control protocol.

### Implications: Scenario One, Null Result

No reactor develops self-replicating order before lifelikeness measurements plateau. The authors interpret this through Gaia: life might require coupled planetary feedback across scales that a laboratory cannot reproduce. Experimental failure alone would not establish that explanation. The scenario identifies dependence on a planetary environment as a possible limit of the apparatus.

### Scenario Two: Great Perceptual Filter

Only the terrestrial simulations produce recognizable life. Cook and Zhu favor a limitation in perception or experimental sophistication over Earth's absolute uniqueness. Their analogy is microbial life before microscopy: failure to observe something may reflect inadequate instruments. A shadow biosphere could remain invisible because the apparatus recognizes only familiar living forms.

### Scenario Three: Cosmoevolutionary Convergence

All successful reactors produce broadly Earth-like biology, including those initialized from exoplanet conditions. The authors compare this with carcinisation, the repeated evolution of crab-like forms. They also suggest panspermia, the distribution of life from a shared ancestor. The essay does not explain how this laboratory outcome would distinguish convergent dynamics from a historical claim about cosmic ancestry.

### Scenario Four: Simulated Contact

Exoplanet-constrained experiments produce unfamiliar lifelike structures. The most pointed example is a convection cell that could detect, process, and consume thermal information to reproduce. Such a phenomenon would escape a detector restricted to chemical composition. The authors also allow that simulated alien life might differ from any life actually inhabiting the modeled planet. Laboratory production would constitute contact with a new possibility, without establishing that an extraterrestrial counterpart exists.

### Scenario Five: Algorithmic Life

Lifelikeness appears in the coordinating algorithms rather than the reactor contents. Selection across several criteria and boxes might produce a collective process that learns how to evolve. Cook and Zhu connect this possibility to artificial general intelligence: discovering the mechanisms of open-ended evolution could change how neural networks train and reproduce. This remains a conjectured consequence. Their closing remarks leave both alternatives open: life might exceed biological substrates, or it might be inseparable from its planetary environment.

## Key concepts

- **Open-ended evolution** — Continued generation of adaptive novelty or complexity. The proposal seeks its mechanisms while acknowledging uncertainty about how to measure it.
- **[Major evolutionary transition](https://antikythera.wiki/terms/major-evolutionary-transition)** — A change in individuality and information processing that alters the possibilities available to subsequent evolution. Machinic language is proposed as a candidate transition.
- **Chemical-computational array** — Parallel physical reactors coordinated by algorithms, with chemical measurements informing changes to experimental conditions.
- **[Assembly theory](https://antikythera.wiki/terms/assembly-theory)** — The source of a proposed molecular complexity measure. Xenoplex would combine that measure with replication and diversity rather than treat it as its only criterion.
- **Great perceptual filter** — The possibility that unfamiliar life remains undetected because instruments and concepts recognize only familiar biology.
- **Simulated contact** — Encountering experimentally generated lifelike structures constrained by exoplanet observations, without assuming they reproduce organisms actually living elsewhere.

## Connections

- [Recursive Worlds](https://antikythera.wiki/work/journal/recursiveworlds) develops Walker's account of life as information structuring matter through evolutionary history. It agrees with Xenoplex's continuity between biosphere and technosphere and gives a fuller treatment of the assembly framework proposed for the reactors. The later work is a conceptual companion, not a reported outcome of Xenoplex.
- [The Long L](https://antikythera.wiki/work/journal/longl) likewise seeks biosignatures that do not presuppose terrestrial biology and describes technological agents through their relations with environments. Its substrate-agnostic ecology supports the project's broadened definition of possible life, while the null-result scenario emphasizes how difficult it may be to isolate life from planetary relations.
- [Substrates Unbound](https://antikythera.wiki/work/journal/substrates) agrees that materials actively shape technical functions. Tripaldi's account sharpens the reason for using actual chemical reactions: their contribution cannot be exhausted by a model chosen in advance. Her attention to material specificity also qualifies any reading of algorithmic life that would make its physical support irrelevant.


## Related works

- [Recursive Worlds](https://antikythera.wiki/work/journal/recursiveworlds)
- [The Long L](https://antikythera.wiki/work/journal/longl)
- [Substrates Unbound](https://antikythera.wiki/work/journal/substrates)