---
title: >-
  MIT randomized professional-writing study: ChatGPT-assisted drafting with human
  submission
slug: chatgpt-professional-writing-rct
stable_id: d06b7c157117b939
company: MIT randomized professional-writing study
function_code: engineering
pattern_codes:
  - autonomous_with_backstop
  - queue_elimination
  - creator_to_judge
evidence_strength: verified
publication_tier: showcase
freshness: current
reviewed_at: '2026-08-23'
updated_at: '2026-08-22'
source_quality_summary: 1 independent, 1 peer reviewed; publication outcomes are verified.
caveat_summary: >-
  Short incentivized tasks, GPT-3.5, and experimental rather than embedded enterprise
  work; no measurement of confidential-data risk.
collections:
  - queue-eliminated
  - embodied-work
bundle_version: 1.0.0
bundle_fingerprint: sha256:c23c6cc2b88153d008ea8fda928f632ce0011fc2d4c5036672a16e5d895bab93
canonical_url: https://brianletort.ai/transformations/chatgpt-professional-writing-rct
---

# MIT randomized professional-writing study: ChatGPT-assisted drafting with human submission

The redesign is a shorter drafting loop with human revision and accountability, not autonomous publication.

Function: Engineering design. Patterns: Autonomous with a backstop; Queue elimination; Creator to judge. Evidence: verified.

Freshness: current. Reviewed: 2026-08-23. Updated: 2026-08-22.


Source quality: 1 independent, 1 peer reviewed; publication outcomes are verified.

## Before

1. **College-educated professional** — Drafts a press release, report, analysis plan, or sensitive email from an occupation-specific prompt. (control: Incentivized task; control average 27 minutes.)
2. **Professional** — Edits and submits the final response. (control: Human owns submission; evaluators grade quality anonymously.)

## After

1. **Professional using ChatGPT** — Prompts the model for a draft or revision during the same occupation-specific task. (control: Tool may generate text, but participant decides whether and how to use it.)
2. **Professional** — Checks, edits, and submits the final work under their own responsibility. (control: Human submission and blind evaluation; experiment did not measure enterprise confidentiality or factual-error incidents.)

## Decision rights

The model proposes text; the professional owns prompts, edits, factual review, and final submission.

## Exception path

Unsuitable, inaccurate, or sensitive text is rejected or rewritten by the professional; the study did not quantify this failure queue.

## Outcomes

- **Task time and evaluator-rated quality** (verified): Control average 27 minutes → Average time decreased 40%; output quality increased 18%. Short incentivized tasks, GPT-3.5, and experimental rather than embedded enterprise work; no measurement of confidential-data risk.

## Executive lesson

Writing acceleration is defensible only when the accountable author owns verification and submission.

## Anti-pattern

Do not project a short-task experiment directly onto confidential or high-stakes enterprise writing.

## Questions for leaders

- Where is the operating threshold set and who can override it?
- What measured result would trigger rollback or retraining?
- Which residual decisions must remain human-owned?

## Sources

- [Experimental evidence on the productivity effects of generative artificial intelligence](https://www.science.org/doi/10.1126/science.adh2586) — Science
- [Experimental Evidence on the Productivity Effects of Generative Artificial Intelligence](https://conference.nber.org/conf_papers/f184551.pdf) — NBER conference paper