CITATION — REFERENCE ENTRY
From Technical Debt to Cognitive and Intent Debt: Rethinking Software Health in the Age of AI — arXiv
- Key
- storey-2026-triple-debt
- Authors
- Storey, Margaret-Anne
- Issued
- 2026-3-23
- Type
- article
- Container
- arXiv
Raw CSL JSON
{
"DOI": "10.48550/arXiv.2603.22106",
"URL": "https://arxiv.org/abs/2603.22106",
"type": "article",
"genre": "Preprint",
"title": "From Technical Debt to Cognitive and Intent Debt: Rethinking Software Health in the Age of AI",
"author": [
{
"given": "Margaret-Anne",
"family": "Storey"
}
],
"issued": {
"date-parts": [
[
2026,
3,
23
]
]
},
"number": "2603.22106",
"container-title": "arXiv"
}
Claims
-
Storey argues that cognitive debt is not new, since developers have always worked with incomplete understanding, but that AI changes how fast it builds up and how hard it is to detect: writing code by hand forces a developer to build at least a partial mental model, while accepting AI-generated code may not.
"Cognitive debt is not new; developers have long worked with incomplete and distributed understanding of complex systems. What is new is the rate at which this gap can accumulate, and the difficulty of detecting it in AI-assisted development. [...] When a developer writes code from scratch, even messy code, the friction and effort mean they build at least a partial mental model along the way. [...] When an AI generates that same code, the developer may accept it without building the same level of understanding."
-
Storey lists signals of cognitive debt, including resistance to change, unexpected results from changes, slow onboarding, loss of knowledge of who knows what, and a low bus factor. She proposes mitigations including code review, pair programming, system walkthroughs, retrospectives, deliberate communication, and having an agent reimplement features to rebuild understanding.
"Resistance to change: low confidence in understanding the system may make developers across the team reluctant to modify it [...] Unexpected results: a reliable signal of cognitive debt is when a team member makes a change expecting one set of observable outcomes, but sees something else entirely [...] Low bus factor: only one person or very few on the team truly understands the system [...] human code review, for example, is valuable not only for catching defects but for spreading understanding across the team. Pair programming serves a similar function. [...] system walkthroughs where developers explain code they did not write [...] reimplementation to repair cognitive debt"
-
Storey describes a student team in her entrepreneurship course whose progress stalled by week eight; the team first blamed technical debt, but the deeper problem was that no one could explain why design decisions had been made or how parts of the system fit together.
"By week eight, one team hit a wall. Simple changes were breaking things in unexpected places, and progress had stalled. When I met with them, they initially blamed technical debt: messy code, hurried implementations, architectural shortcuts. But as we dug deeper, a different problem emerged. No one on the team could explain why certain design decisions had been made, or how different parts of the system were supposed to work together."
-
Storey identifies cognitive surrender, as described by Shaw and Nave, as the psychological mechanism behind cognitive debt, and argues that because surrender inflates confidence, it helps explain why teams do not notice cognitive debt until it is too late.
"The psychological mechanism behind this is what Shaw and Nave [2026] refer to as cognitive surrender: adopting AI outputs with minimal scrutiny [...] Notably, Shaw and Nave find that cognitive surrender also inflates confidence even when the AI is wrong, which helps explain why cognitive debt remains invisible until it is too late: the team feels they understand the system better than they do."
-
Storey defines cognitive debt as a team-level, project-level property: the erosion of shared understanding of a software system over time. She notes that the term has also been used for reductions in individual neural engagement during AI-assisted tasks, but says her use focuses on the team and on change over time.
"Cognitive debt is a team-level, project-level property reflecting the erosion of shared understanding across a software system over time. [...] The term has also been used to describe measurable reductions in individual neural engagement during AI-assisted tasks [...] Our use of the term, however, focuses on the team-level and longitudinal dimension of software development: the accumulated erosion of shared understanding of a software system over time."
-
Storey proposes a triple debt model: technical debt concerns problems in code, cognitive debt the erosion of a team's shared understanding over time, and intent debt the lack of externalized goals, constraints, and rationale. Technical debt makes systems harder to change, cognitive debt harder to understand, and intent debt harder to know the purpose of.
"technical debt refers to problems in the code layer, cognitive debt refers to erosion of shared understanding across a team over time, and intent debt refers to a lack of externalized goals, constraints, and rationale that both humans and AI systems need to work safely and efficiently with the codebase. Technical debt makes systems harder to change. Cognitive debt makes systems harder to understand. Intent debt makes it difficult to know what the system is actually for."
Available in