On the Foundations of Trustworthy Artificial Intelligence

TL;DR

This paper establishes that platform-deterministic inference is essential for trustworthy AI, introduces trust entropy to measure non-determinism costs, and demonstrates a practical, cross-architecture deterministic inference system verified through extensive testing.

Contribution

It formalizes the Determinism Thesis, introduces trust entropy, and develops a cross-architecture deterministic inference engine to ensure AI trustworthiness.

Findings

Verification probability equals 1 - 2^{-H_T}

Zero hash mismatches in 82 tests across architectures

Identical outputs verified across distributed nodes

Abstract

We prove that platform-deterministic inference is necessary and sufficient for trustworthy AI. We formalize this as the Determinism Thesis and introduce trust entropy to quantify the cost of non-determinism, proving that verification failure probability equals 1 - 2^{-H_T} exactly. We prove a Determinism-Verification Collapse: verification under determinism requires O(1) hash comparison; without it, the verifier faces an intractable membership problem. IEEE 754 floating-point arithmetic fundamentally violates the determinism requirement. We resolve this by constructing a pure integer inference engine that achieves bitwise identical output across ARM and x86. In 82 cross-architecture tests on models up to 6.7B parameters, we observe zero hash mismatches. Four geographically distributed nodes produce identical outputs, verified by 356 on-chain attestation transactions. Every major trust property of AI systems (fairness, robustness, privacy, safety, alignment) presupposes platform determinism. Our system, 99,000 lines of Rust deployed across three continents, establishes that AI trust is a question of arithmetic.