# The Augmentation Trap: AI Productivity and the Cost of Cognitive Offloading

arXiv: https://arxiv.org/abs/2604.03501 · HTML: https://mlsanity.com/paper/2604.03501/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 48 references that are in the map.
- First posted on arXiv 3 Apr 2026.

## Builds on (most cited first, 8 of 8)

- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [The Impact of AI on Developer Productivity: Evidence from GitHub Copilot](https://mlsanity.com/paper/2302.06590/) (2023)
- [When combinations of humans and AI are useful: A systematic review and meta-analysis](https://mlsanity.com/paper/2405.06087/) (2024)
- [Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task](https://mlsanity.com/paper/2506.08872/) (2025)
- [How AI Impacts Skill Formation](https://mlsanity.com/paper/2601.20245/) (2026)
- [Collaborating with AI Agents: Field Experiments on Teamwork, Productivity, and Performance](https://mlsanity.com/paper/2503.18238/) (2025)
- [From Future of Work to Future of Workers: Addressing Asymptomatic AI Harms for Dignified Human-AI Interaction](https://mlsanity.com/paper/2601.21920/) (2026)
- [Personality pairing improves human-AI collaboration](https://mlsanity.com/paper/2511.13979/) (2025)

## Built on by (most cited first, 4 of 4)

- [Human-AI Productivity Paradoxes: Modeling the Interplay of Skill, Effort, and AI Assistance](https://mlsanity.com/paper/2605.11350/) (2026)
- [Exploratory Responsiveness and Adaptive Rigidity under AI-Assisted Optimization](https://mlsanity.com/paper/2606.10086/) (2026)
- [Predictive Assistance and the Temporal Dynamics of Exploratory Compression](https://mlsanity.com/paper/2606.10094/) (2026)
- [Managing the Human Fallback: Skill Investment Under Improving AI and Worker Mobility](https://mlsanity.com/paper/2606.29111/) (2026)

## Abstract (arXiv)

Experimental evidence suggests that AI tools raise worker productivity, but also that sustained offloading can erode expertise. This creates a tradeoff when expertise is a complement to AI. To explore the consequences of this tradeoff, we develop a dynamic model in which a decision-maker chooses the intensity of practice-displacing AI offloading for a worker over time, trading immediate productivity against the erosion of worker skill. We decompose the tool's productivity effect into two components, one independent of worker expertise and one that scales with it. The model produces three main results. First, a decision-maker who fully anticipates skill erosion still rationally adopts AI when front-loaded gains outweigh long-run skill costs, lowering long-run productivity. The decomposition sorts deployments into five regions by their long-run effect, separating beneficial from harmful adoption. Second, the tradeoff introduces the potential for misaligned incentives to harm workers. When managers are short-termist or worker skill has external value, AI use can leave the worker worse off than with no AI, the outcome we call the augmentation trap. Third, when AI substitutes strongly enough for expertise, offloading can generate permanent divergence, with high-skill workers realizing their potential and low-skill workers deskilling. Small differences in managerial incentives can determine which path a worker takes.