Drafting recommendation letters that stay honest and specific
Every fall brings a stack of recommendation letters, each needing an hour you don't have — and Anthropic's data shows drafting them is one of the admin tasks professors most often hand to AI. Done carelessly, that means template letters and student data in a consumer tool. Done right, it means your notes, your judgment, and a draft that saves the composition time.
You are helping a professor draft a letter of recommendation. Privacy rules for this chat: I will refer to the person only as "the student" — never ask for a name — and you will write the letter with [Name] placeholders that I will fill in later in my word processor. My notes on the student: {{student_notes}} They are applying to: {{target_program}} How I know them: {{relationship_context}} Endorsement strength, honestly stated: {{endorsement_level}} Structure: 1. Opening: who I am, how long and in what capacity I have known [Name], and a one-sentence endorsement that matches the strength I stated 2. Two body paragraphs, each built around one concrete example from my notes — evidence, not adjectives 3. A closing that connects [Name]'s record to this specific program and offers contact for follow-up Constraints: use only facts from my notes — do not invent projects, grades, rankings, or personality traits. No stock phrases ("It is my distinct pleasure..."). Do not escalate my stated endorsement level; a credible letter serves the student better than an inflated one. If my notes are too thin to support a persuasive letter, reply "NEED MORE" and ask me three specific questions instead of padding.
Fill in your details and the prompt updates live — then copy.
Dear Members of the Admissions Committee, I write to give [Name] my strong recommendation for your doctoral program. I have known [Name] for three years, first as the instructor for two courses and then as [Name]'s independent-study supervisor, and I place [Name] among the five strongest students I have supervised in the past decade. The clearest evidence of [Name]'s promise came during an independent replication of a published labor economics study. Rather than stopping at reproducing the headline results, [Name] traced a discrepancy to a coding error in the original analysis — the kind of care that distinguishes...
The full workflow
- Confirm the student's consent and any waiver paperwork first, as you would for any letter.
- Write five minutes of honest shorthand notes with concrete examples; the letter can only be as specific as your notes.
- Generate the draft, then add the name, letterhead, and any detail only you would know — in your word processor, never in the chat.
- Read it aloud once; if it could describe any strong student, it isn't done.
Watch out for
FERPA: recommendation letters draw on education records. Keep the student's name, grades-plus-identity, and application materials out of consumer AI tools — draft with placeholders and add identifying details only in your own document. If your institution licenses a protected tool (e.g., Copilot under an enterprise agreement), prefer it.
Admissions committees read thousands of letters and increasingly recognize AI boilerplate. A detectable template letter damages your credibility and the student's application together.
Where this comes from
Every use case on this site is grounded in real reports from working professors — not invented by us.