The Productivity-Reliability Paradox: Specification-Driven Governance for AI-Augmented Software Development
Sabry E. Farrag · 2026
Read The Productivity-Reliability Paradox: Specification-Driven Governance for AI-Augmented Software Development by Sabry E. Farrag (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-01160.txt.
Description
Sabry E. Farrag argues that AI coding assistants create a Productivity-Reliability Paradox in software development. Trials report large output gains on well-scoped tasks, while a careful randomized trial records a slowdown for experienced coders. Large telemetry shows many more pull requests but much longer review times with flat delivery. Farrag defines the paradox in formal terms through a multivocal review of dozens of sources.
Three moderating variables and two amplifying mechanisms structure it. A scheme sorts six methodologies under three AI adoption tiers, and a Specification Governance Model grounded in transaction cost economics offers a practical decision guide. Two tools are tested as model examples in a four-month pilot. Specification discipline, not model capability, is named the binding constraint.
Catalog record
| CREATOR | Sabry E. Farrag |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 90 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.01160