AI & Decisions · 2 minute read
Why an AI's First Answer Can Shape Your Judgment
An AI estimate can become an anchor before you have formed your own view, making later reasoning orbit around the model's first number.
A normal situation
You ask an AI how long a migration should take and it answers six weeks. You then adjust the number to seven or eight weeks, but never build an estimate that is independent of the original six.
What is happening?
The model's answer is now the starting point. Even careful adjustment can stay too close to that first value because the comparison has already been framed.
AI makes first estimates extremely easy to obtain. That is useful, but it also means an unverified number can enter a decision before anyone has checked a base rate, comparable case or constraint.
The useful lens
Anchoring Effect
This does not prove that one bias caused the decision. It is a useful lens for asking a better question and checking the evidence.
Evidence note: Anchoring is a well-supported effect in which an initial numerical value can pull a later estimate toward it. A 2026 meta-analysis covering 2,601 effect sizes found a large overall effect, but also substantial variation across studies. The effect should not be treated as a rule that every number changes every judgment: incidental anchors, anchors from a different dimension, clearly random values, incentives, and some debiasing conditions were associated with smaller or null effects.