Ars Technica's argument that coding agents generate more code but not necessarily more software is a useful correction to the productivity debate.
Software value comes from working systems, maintainable architecture, tests, security, product judgment, and operational reliability. More generated code can help, but it can also create review burden and technical debt.
The teams that benefit most from coding agents will measure shipped outcomes rather than lines produced. The agent should compress real engineering work, not simply expand the diff.
Was this useful?
Help Pagish understand which AI stories are worth covering more deeply.
Tell Pagish if this story was useful.