The 4 Laws of Probabilistic Software
Four foundational laws governing the behavior, economics, and maintenance of AI-generated code. Law 1: AI code is probabilistic, not deterministic. Law 2: Complexity scales non-linearly with AI assistance. Law 3: The verification cost of AI code exceeds the generation cost. Law 4: AI-generated code accumulates debt faster than human-written code. These laws, coined in Built In, form the baseline for managing modern, AI-augmented engineering teams.
“We have automated the typing, but we have not automated the thinking.”
The industry is treating AI-generated code as a free lunch, assuming that faster code generation strictly equates to higher productivity. The 4 Laws establish that the physics of software engineering have changed. Because the code is probabilistic, it introduces subtle, compounding errors that require massive human oversight. Ignoring these laws leads directly to the negative-carry code crisis, where systems become unmaintainable due to the sheer volume of unverified, machine-generated complexity.
Multi-Hop Causal Traversal Engine
Explore how concepts dynamically feed into each other across 1-hop, 2-hop, and 3-hop transitive relationships. Click any node to navigate the causal highway.
The 4 Laws of Probabilistic Software
Four foundational laws governing the behavior, economics, and maintenance of AI-generated code. Law 1: AI code is probabilistic, not deterministic. Law 2: Complexity scales non-linearly with AI assistance. Law 3: The verification cost of AI code exceeds the generation cost. Law 4: AI-generated code accumulates debt faster than human-written code. These laws, coined in Built In, form the baseline for managing modern, AI-augmented engineering teams.
Direct Relationships (6)
Transitive Neighbors (Connected via Hop 1)
Extended Causal Ripple Effects
Richard Ewing’s Research Thesis
Embrace AI generation, but govern it with ruthless, deterministic verification.
Why This Specification Exists
Engineering teams are blindly accepting AI-generated code and accumulating massive technical debt.
Treating AI tools as standard IDE autocomplete features.
No fundamental principles defining the unique economic and structural reality of probabilistic code.
Four laws that clarify the necessary operational shift for AI-augmented teams.
What Changes If You Believe This?
Processes shift focus from writing code to reading, reviewing, and testing code.
Accounts for the verification tax when forecasting engineering bandwidth.
Adjusts release expectations acknowledging the verification bottleneck.
Implements stricter scanning on AI-generated pull requests.
Recommended Action by Role
Increase time allocated for code reviews to account for Law 3.
Latest Publications & Research Activity
The Engineering Bottleneck Illusion: What Copilot Adoption Taught Us
Cursor vs Google Antigravity for Production AI Building
Most Companies Shouldn’t Be Using Autonomous Coding Agents Yet
Frequently Asked Questions
Q:Why does AI code accumulate debt faster?
AI often generates verbose, locally-optimized code that lacks systemic architectural awareness, leading to fragmentation and duplicate logic over time.
Q:What does probabilistic mean in this context?
It means the same prompt can yield different code on different days, removing the predictable, mechanical certainty traditional engineers rely upon.
Canonical Specification Origin
Embrace AI generation, but govern it with ruthless, deterministic verification.
Corpus Interconnections
Richard Ewing artifacts developed around this canonical framework, including publications, execution tools, and diagnostic models.
External Adoption & Peer Citations
Documented instances where independent researchers, engineering teams, and publications have cited, implemented, or referenced this concept outside Richard Ewing’s ecosystem.
External Evidence: No independently verified references recorded yet.
This concept is part of Richard Ewing’s original baseline canon. External citations and implementations are added only upon rigorous empirical verification.
Inspectable Evidence Ledger
Classified evidence items supporting, extending, or refining this canonical research specification.
| Evidence Item | Publisher | Evidence Type | Strength | Role | Action |
|---|---|---|---|---|---|
| The Negative-Carry Code Crisis | Beehiiv | Newsletter | ★★★★★ | Origin | Inspect ↗ |
| The Vibe Coding Era | Built In | Industry Article | ★★★★ | Supports | Inspect ↗ |
| The Copilot Bottleneck | CIO.com | Tier-1 Article | ★★★★★ | Extends | Inspect ↗ |
| Model Collapse | CIO.com | Tier-1 Article | ★★★★ | Extends | Inspect ↗ |
| The Engineering Bottleneck Illusion: What Copilot Adoption Taught Us | Newsletter | ★★★★★ | Extends | Inspect ↗ |
Recommended Citation
Ewing, R. (2026). "The 4 Laws of Probabilistic Software." Richard Ewing Research Canon. Available at: https://www.richardewing.io/concepts/four-laws-probabilistic-software
@article{ewing_four_laws_probabilistic_software,
author = {Ewing, Richard},
title = {The 4 Laws of Probabilistic Software},
journal = {Richard Ewing Research Canon},
year = {2026},
url = {https://www.richardewing.io/concepts/four-laws-probabilistic-software}
}