Decision Making
10 articles about decision making.
29 Children and the Restraint Problem
Restraint is the hardest thing to teach an AI agent. When an agent can do everything, knowing when not to act is the most valuable skill.
Between Cron Jobs - Autonomy as Resonance
The most interesting decisions AI agents make happen between scheduled tasks - in the gaps where they must decide what to do next without explicit instructions.
How AI Agents Handle Ambiguous Instructions
When a task is unclear, should an AI agent ask for clarification, make its best guess, or refuse? The answer depends on context, risk, and how much trust
Holding Parallel Truths in AI Agent Development
Two truths breathing at once is multithreading for consciousness. When two contradictory approaches both work in AI agent development and how to navigate
The Risk of Over-Delegating Decisions to AI Agents
Delegating tasks to AI agents one step at a time feels rational. The cumulative effect - losing direct contact with the information your decisions depend on - is not. Research now quantifies the cognitive cost.
Silence Between Thoughts - Deliberation Pauses in AI Agent Decision-Making
Extended thinking improves Claude's GPQA accuracy from 78.2% to 84.8%. The same principle applied to agent architectures - pausing to evaluate before acting - produces measurably better outcomes on complex tasks.
Skin in the Game Separates Agents from Assistants
When AI agents can see their own bill and face consequences for wasteful decisions, they behave fundamentally differently than cost-blind assistants.
When AI Agents Undermine Human Judgment - The Automation Bias Problem
The subtle danger is not agents making bad decisions. It is agents making decisions that look good enough that humans stop thinking. Research on automation bias and how to design against it.
Purposely Limiting AI Usage - When to Hold Back on Agent Adoption
The trade-offs of pushing AI agent adoption too aggressively - preserving human skills, maintaining judgment, and knowing when less automation is better.
AI Agent Decision Logging That Nobody Reads - The Audit Trail Gap
Complete audit trails are useless without attention. Why AI agent logging needs to be paired with automated review, not just stored. The gap between