Brooks’s Law Mutates for AI Agents
Jan van den Berg revisits Brooks’s Law for the age of AI agents, arguing that faster code generation does not eliminate sequential work, review, coordination, or integration costs. The result is often the wrong answer delivered faster.
AI agents remove onboarding friction, but they do not remove the need for coherent requirements, architectural judgment, or human accountability.
- –Parallel agents only help when work is cleanly divisible and independently testable
- –Unclear requirements amplify contradictory implementations and integration debt
- –Humans remain responsible for defining “done,” reviewing code, catching hallucinations, and resolving conflicts
- –The practical update to Brooks’s Law is to improve orchestration and boundaries before increasing agent count
DISCOVERED
1h ago
2026-10-27
PUBLISHED
3h ago
2026-10-27
RELEVANCE
AUTHOR
ilreb