About

The Intelligent Commit looks at what AI is doing to the way software is built, reviewed, secured and run. It is written for engineers and engineering leaders who have to decide how far to trust AI tools, and then live with the results.

Most of the coverage is about what goes wrong. It asks where AI-generated code fails, what the failure costs and who is accountable when an agent gets something wrong.

What's here

• Opinions are longer pieces on where AI is taking engineering teams.
• Research takes one recent paper at a time and separates what it measured from what it means for the people who build software.
• The Complexity Paradox is a series on the security of LLM-generated code, built on a study in which the simple tasks produced nine times more vulnerabilities per line than the complex ones.
• The news log records each week's AI industry stories, filed under ten topics.
• Statistics is a library of figures on AI in software development, each linked to its source.

New opinion and research pieces go out by email as they're published. Subscribe to have them sent to your inbox.