Intel / article

(Im)Paired Programming: Coding Agents Improve Productivity but Harm Understanding

Source: arxiv.org
Author: Nishant Balepur, Connor Baumler, Valerie Chen, Eunsol Choi, Rachel Rudinger, Jordan Lee Boyd-Graber
Date: Published Jul 30, 2026

Context

Finishing a coding task with an did not ensure that students understood the resulting code. In a study of 54 students building websites, researchers compared an agent that edited code with a chatbot that left students to write or adapt code themselves. Agent assistance helped initial completion, but students performed worse on comprehension questions and were less prepared to extend their code without an agent. Copy-pasted prompts and automatically accepted edits were also associated with weaker comprehension. For learning and oversight, the study identifies a gap between obtaining working software and being able to explain or modify it independently. Study abstract.

Key takeaways · AI-distilled

54 students built a website with either a coding agent that edits their code or a chatbot they had to copy from. The agent group finished better but scored worse on comprehension questions — and then could not extend their own code once the agent was taken away.
The damage tracks how you interact, not just whether you use an agent: copy-pasting prompts and auto-accepting edits are the specific behaviors linked to lower comprehension. Effort spent reading and steering is what preserves understanding.
Users kept preferring the agent even after reporting they understood their code less. Speed and ease win the preference vote, so you cannot rely on developers self-selecting away from the tool that is quietly costing them.
The suggested fixes are product-side, not willpower-side: discourage low-effort prompting, make agents emit readable code, and design interactions that force active engagement — because 'human in the loop' means nothing if the human stopped understanding.

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.

Key quotes

“Coding agents (e.g., Cursor) improve developer productivity by optimizing task completion, but shifting users from writing code to prompting and reviewing may harm their understanding, impeding oversight, learning, and communication.”

“While agents aid initial task completion, they harm users' code comprehension and thus do not prepare users to extend their code;”

“Despite self-reported weaker understanding, users still prefer coding agents because they are quick and easy to use.”

“While users stay in the loop for coding workflows, understanding should not be forgotten.”

Read the source arxiv.org