The First Draft Trap: When AI Quietly Hijacks Legal Judgment
Lawyers have a new productivity tool that may be quietly doing their thinking for them. The moment you accept an AI-generated draft as your starting point, you have already made the most consequential decision of the entire project - by not making it deliberately. Everything that follows is editing, not thinking.
This is the First Draft Trap, and it inverts the core training of the legal profession.
How anchoring distorts judgment
Cognitive psychologists Daniel Kahneman and Amos Tversky demonstrated decades ago that once people encounter an idea, that first impression distorts all subsequent judgments. In their experiments, subjects who watched a roulette wheel land on a random number still let that number influence their estimates of completely unrelated quantities. The anchor held even when people knew it was meaningless.
An AI-generated draft is the most seductive anchor imaginable. It is not random - it is plausible, organized, and sounds like something a lawyer would write. You know intellectually it is just one of many possible approaches. The anchor holds anyway.
The AI draft becomes a filter that prevents you from seeing other options. You never even notice the roads you did not take.
The opposite of how lawyers are trained
From the first day of law school, the Socratic method teaches lawyers to resist the obvious answer. A professor hears a confident response and asks: What else? What if the facts were different? What is the argument on the other side?
The goal is to build the mental habit of holding multiple possibilities in tension before committing to any one of them. The First Draft Trap delivers a confident answer before you have even formulated the question properly.
Consider what a senior partner actually does when reviewing a junior associate's memo. The partner's value is not better writing. It is peripheral vision - the ability to see what the memo does not address, the argument not considered, or the framing that would land differently with a particular judge or jury. That capacity atrophies when your default workflow begins with a prompt.
System 1 thinking masquerading as System 2
Kahneman and Tversky's two-system framework explains what is happening. System 1 is fast, intuitive, and pattern-matching. System 2 is slow, deliberate, and analytical. The practice of law, at its best, is a System 2 discipline.
An AI draft feels like a System 2 output. It is structured, footnoted, methodical. But your decision to accept it as a starting point is pure System 1 - a fast, intuitive grab at the nearest plausible answer. You have used a sophisticated tool to bypass the sophisticated thinking the tool was supposed to support.
That uncomfortable period of ambiguity, of not knowing which path is best, is where the real lawyering lives.
Ask for the map, not the draft
Stop using AI to skip the hard part that matters. Before you ask for a draft, ask for the map.
Describe the matter or document you are working on, then request three fundamentally different strategic framings for the problem. For each framing, ask the AI for the strongest argument in its favor and its most serious vulnerability. Then ask which framing best fits the client's goals, the audience, or the procedural posture.
Close with a clear instruction: Do not write a draft yet.
That last instruction is the key. It keeps you in the driver's seat during the phase that matters most. You use AI to expand the possibilities before you prune them, not after. You retain the opportunity to think for yourself about other important possibilities and add them in.
Effective prompt engineering means structuring requests to generate options rather than conclusions. Let the machine generate possibilities. You exercise judgment.
For lawyers, the ability to see what is not there is the whole game. Do not let the first draft blind you to it.
Learn more: Explore how AI tools should be used in legal practice to support rather than replace critical thinking.
Your membership also unlocks:
