---
title: 'The Negentropy Substrate: A First-Principles Validation of Nature''s Intelligence as the Training Ground for Physically-Grounded AGI'
slug: 'negentropy-substrate-agi'
date: 2026-05-13
type: 'essay'
status: 'published'
tags: ['enviroai', 'ai', 'information-theory', 'paper']
abstract: 'Argues that the next paradigm of AI development must move beyond the statistical scaling of language models toward physically-grounded intelligence trained on nature''s own negentropic algorithms.'
license: 'CC-BY-4.0'
author: 'Jed Anderson'
co_authors: []
canonical_url: 'https://jedanderson.org/essays/negentropy-substrate-agi'
original_date: 2025-11-20
pdf: '/pdfs/negentropy-substrate-agi.pdf'
hero_image: '/images/negentropy-substrate-agi-hero.jpg'
hero_image_alt: 'Cover page of The Negentropy Substrate: A First-Principles Validation of Nature''s Intelligence as the Training Ground for Physically-Grounded AGI'
supporting_files: []
---

Argues that the next paradigm of AI development must move beyond the statistical scaling of language models toward physically-grounded intelligence trained on nature's own negentropic algorithms.

*Full text in the canonical PDF above; markdown body pending extraction during review.*