Lines of code got a better publicist
curlewis.co.nz
June 11, 2026
7 min read
🔥🔥🔥🔥🔥
62/100
Summary
Measuring developer productivity solely by lines of code is misleading and does not reflect their true impact on business outcomes. Evaluating contributions should focus on what features were delivered, their benefits to customers, and their effect on revenue and reliability.
Key Takeaways
- Google reports that 75% of new code is AI-generated, while Anthropic claims that around 80% of merged production code is written by its AI, Claude.
- A survey of approximately 6,000 executives found that 69% of firms using AI reported no measurable productivity impact, with a consensus indicating around 10% organizational gains.
- The industry has shifted from outcome-based metrics to volume claims regarding AI-generated code, which may not correlate with improved developer productivity or customer satisfaction.
- Research indicates that while AI may speed up development processes, measuring its impact on productivity has become increasingly complex and unreliable.
Community Sentiment
Positives
- The shift towards more pragmatic discussions about code generation reflects a growing awareness of the importance of maintainability over sheer volume, which could lead to better software practices.
- The term 'slop' effectively communicates the concerns around unmanageable code generated by AI, highlighting the need for quality over quantity in AI applications.
Concerns
- The emphasis on lines of code as a measure of productivity undermines the complexity of software development, risking a return to outdated metrics that do not reflect true quality.
- Claims about AI-generated code often lack transparency and verifiable value, raising concerns about the actual utility of such outputs compared to traditional coding practices.
- The pressure to adopt AI rapidly, without thorough evaluation of its claims, may lead to significant technical debt and unmaintainable codebases, echoing past mistakes in software engineering.
