Related video: How AI Accelerators, GPUs and CPUs Think Differently, @IBMTechnology
Trending on Hacker NewsHarvard study: AI coding agents speed output but slow code review
complexdiscovery.com · · 25-second read
Top comments from YouTube.
This story is from November 17, 2026. For the latest, see today's news.
November 17, 2026 · AI Tech
Harvard researchers studied how AI coding agents affect software development workflows. They found that increased code output by 30 percent compared to human developers working alone. However, the finished work completed did not rise measurably, suggesting efficiency gains disappeared elsewhere in the pipeline.
The critical finding: code review time increased by 49 percent. The study suggests AI-generated code, while faster to produce, requires significantly more scrutiny and revision before it can be merged into production systems. This shifts the development bottleneck from initial coding to the review and validation stage.
This is an AI-written summary. The original reporting belongs to complexdiscovery.com. How we write these · Spot a mistake?
Get the AI brief every morning.
The five AI stories that matter each morning: startups, money, jobs, tech and what's coming next. Free, and one click to unsubscribe. See our Privacy Policy.