Thinking Clearly: Key Ideas from the Video

Always seek consensus before acting

Adopt a first-principles, evidence-seeking mindset

Rely mainly on intuition and gut feeling

Defer all decisions to authority figures

Make decisions only with unanimous team approval

Double down on initial judgments to show confidence

Break problems into smaller, testable parts

Postpone any action until perfect information emerges

Ignore incentives if someone seems credible

Treat incentives as irrelevant to outcomes

Look for how incentives shape behavior rather than assume goodwill

Assume people act solely from moral intentions

Rely on single-case vivid examples

Use base rates and reference-class forecasting

Forecast based on personal optimism alone

Increase confidence by ignoring past failures

Outcomes become fully predictable and stable

Everyone naturally aligns with the system's long-term goals

Motivation always increases without incentives

People optimize for unintended metrics and create perverse outcomes

Repeat the same approach until it works

Conduct rapid, low-cost experiments and update beliefs

Hide mistakes to protect reputation

Wait for others to point out errors

Experts always outperform simple statistical rules

Expert predictions are always reliable without calibration

Only novices make prediction errors

Experience never improves judgment

Ask what observable test would prove or disprove it soon

Approve plans only if they are risk-free

Favor plans that sound ambitious regardless of evidence

Choose the plan that maximizes short-term optics