GitHub - psyll/Epistemic-Heat-Death-and-the-Signal-to-Noise-Ratio-of-the-Global-Web: Explores the structural degradation of the open web caused by large-scale synthetic content production. Introduces a formal framework with three laws, a dynamic ODE model, SNR index, entropy taxonomy, and the transition from SEO to Authority Optimization (AEO) based on cryptographic provenance. · GitHub
/" data-turbo-transient="true" />
Skip to content
Search or jump to...
Search code, repositories, users, issues, pull requests...
-->
Search
Clear
Search syntax tips
Provide feedback
--><br>We read every piece of feedback, and take your input very seriously.
Include my email address so I can be contacted
Cancel
Submit feedback
Saved searches
Use saved searches to filter your results more quickly
-->
Name
Query
To see all available qualifiers, see our documentation.
Cancel
Create saved search
Sign in
/;ref_cta:Sign up;ref_loc:header logged out"}"<br>Sign up
Appearance settings
Resetting focus
You signed in with another tab or window. Reload to refresh your session.<br>You signed out in another tab or window. Reload to refresh your session.<br>You switched accounts on another tab or window. Reload to refresh your session.
Dismiss alert
{{ message }}
psyll
Epistemic-Heat-Death-and-the-Signal-to-Noise-Ratio-of-the-Global-Web
Public
Notifications<br>You must be signed in to change notification settings
Fork
Star
main
BranchesTags
Go to file
CodeOpen more actions menu
Folders and files<br>NameNameLast commit message<br>Last commit date<br>Latest commit
History<br>2 Commits<br>2 Commits
charts
charts
scripts
scripts
.gitattributes
.gitattributes
Epistemic Heat Death and the Signal-to-Noise Ratio of the Global Web.md
Epistemic Heat Death and the Signal-to-Noise Ratio of the Global Web.md
LICENSE
LICENSE
README.md
README.md
View all files
Repository files navigation
Epistemic Heat Death and the Signal-to-Noise Ratio of the Global Web
Author: Jarosław Szulc · psyll.com · jarek@psyll.com
Working paper · v2.0 · July 2026 · Draft for public comment
A theoretical framework for the post-AI information ecology: as AI-generated text, images, audio, and video come to dominate newly published web content, the web can look more abundant than ever while its actual information content - distinct ideas, grounded claims, accountable authorship - quietly collapses. This paper calls that failure mode epistemic heat death and builds a formal apparatus around it: three governing laws of signal value, a coupled-ODE dynamical model, a proposed Signal-to-Noise Ratio (SNR) index, and a technical/economic blueprint (the cryptographic provenance stack) for an alternative equilibrium.
What's in this repo
Epistemic Heat Death and the Signal-to-Noise Ratio of the Global Web.md - the full paper
scripts/ - reference Python implementations for the paper's formal models, one module per section, each runnable standalone
charts/ - figures generated directly from the reference implementations
Scripts → paper sections
Script<br>Paper section<br>Depends on
laws.py<br>§3–4 - the three laws of epistemic signal value, and the epistemic debt equation<br>stdlib
dynamics.py<br>§5 - the coupled-ODE dynamic model, scenario trajectories, sensitivity grid, AEO intervention scenario<br>numpy, scipy
snr_index.py<br>§6 - the three-level SNR index (document / domain / global web) and the EVI metric<br>stdlib
entropy.py<br>§10 - the entropy taxonomy (statistical, semantic, epistemic, authorship) and decay measures<br>stdlib
Each script runs on its own and prints worked examples tied to the paper's numbered results:
pip install numpy scipy<br>python3 scripts/dynamics.py<br>python3 scripts/laws.py<br>python3 scripts/snr_index.py<br>python3 scripts/entropy.py
dynamics.py, for instance, reproduces the baseline trajectory (§5.2), the scenario comparison (§5.3), the sensitivity-grid extremes (§5.4), and the AEO intervention comparison (§5.5) as console output.
Key claims
The web is undergoing a measurable rise in epistemic entropy that, absent intervention, tends toward a degenerate equilibrium - epistemic heat death.
Diversity loss isn't one phenomenon but at least four loosely-coupled ones (statistical, semantic, epistemic, authorship entropy), which can move independently and even in opposite directions.
Pageviews and engagement metrics are structurally broken as value proxies once a substantial share of traffic and content production is non-human.
The same mechanisms extend to images, audio, and video (synthetic reality collapse), with failure modes - chiefly provenance-metadata stripping on re-compression - that text alone doesn't have.
Epistemic heat death is not distributionally neutral: it produces winners and losers along lines of expertise, language, geography, and institutional access, and is exploitable as a tool of state and corporate power.
In simulation, the model's SNR_proxy(t) declines monotonically across the full tested parameter grid; the epistemic half-life is...