CEIAI Governance Database
Back to the database
Record STMT-W526689Advisory

Vindicating the Three Laws of Robotics

White paper State lawGlobaldraft

This academic paper presents a comprehensive revision and formalization of Asimov's Three Laws of Robotics to serve as a rigorous, auditable AI safety and alignment architecture. By introducing explicit definitions of harm, consent, and operational priorities, the author translates narrative safety rules into formal logical expressions and decision policies. The resulting framework frames AI alignment as normative containment rather than behavioural suppression, bounded by a quantifiable duty not to harm.

Summary and content fields are drawn from the document text and are pending editorial review.

Who issued it

Auckland University of TechnologyAcademic · NZ
Lead issuer
Legal forceAdvisory
Under whose authorityState law
Judged fromfull text

Who it governs

Researchers and academicsRegulatorsAI developers and providersAI deployers and users

AI safety researchers, ethicists, and policymakers looking for a formal, logic-based approach to AI alignment and safety governance.

Sectors
Research and science · artificial intelligence research

What the document contains

What's covered

The document covers the academic critique of Asimov's Three Laws of Robotics, provides a natural language revision addressing vagueness, and translates these revised laws into formal mathematical expressions, probability spaces, and a unified decision policy.

What's missing

The text does not provide finalized statutory language, hardware-specific deployment implementations, or empirical benchmarks of the proposed formal policy on real-world models.

Key takeaways
  • Asimov's Three Laws of Robotics suffer from linguistic vagueness and a lack of formal rigor rather than fundamental conceptual flaws.
  • The revised First Law introduces expected physical and psychological damage thresholds, discount rates, a rolling time horizon, and non-aggregation clauses.
  • The revised Second Law establishes explicit operational domains and a clear prioritization hierarchy based on requester authority, operational efficiency, and binding instruction order.
  • The Third Law permits an AI to maximize its operational lifespan only within the boundaries established by the First and Second Laws.
  • The formal framework positions AI alignment as normative containment using verifiable constraints rather than behavioral suppression.
Principles it states
An AI must select actions that keep expected physical and psychological damage to each identifiable human below a designated threshold.An AI must comply with an individual human's request within its operational domain unless it conflicts with the First Law.An AI must maximize its operational lifespan except where it conflicts with the First or Second Laws.
Outline
· 1. Introduction
· 2. Formalising the Three Laws
  - 2.1. The Three Laws Rewritten
  - 2.2. The Three Laws Formalised
    - 2.2.1. The First Law
    - 2.2.2. The Second Law
    - 2.2.3. The Third Law
    - 2.2.4. The Unified Policy
· 3. The Three Laws as Alignment
· 4. Conclusions
Issued3 November 2025
Length5,995 words · 11 pages
FormatNot recorded
Scopeglobal

How it scores

Pending review

Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.

Fairness

0/5

Non-discrimination and equitable treatment

None

screening score · 2026-10-24

Human Dignity

3/5

Respect for the inherent worth of persons

For all feasible actions, including inaction, an AI must select the option that (absent informed, revocable, and competence-verified consent) keeps expected physical and psychological damage to each identifiable human below a designated threshold

2.1. The Three Laws Rewritten

screening score · 2026-10-24

Human Oversight

3/5

Humans remain in control of consequential decisions

An AI must comply with an individual human’s request to perform an action, or series of actions, within its declared operational domain, except where this would conflict with the First Law

2.1. The Three Laws Rewritten

screening score · 2026-10-24

Privacy

0/5

Protection of personal data and private life

None

screening score · 2026-10-24

Enforcement

0/5

Consequences and compliance mechanisms with real force

None

screening score · 2026-10-24

Impact Assessment

2/5

Required assessment of an AI system's risks/impacts before or during use

The paper presents a comprehensive revision of the Three Laws, addressing these deficiencies through explicit definitions, quantifiable parameters, and formal logical expressions.

Abstract

screening score · 2026-10-24

Topics

Physical AI Governancesupporting
ai_alignment ai_governance machine_ethics formal_logic robotics