Skip to content
Slow Variables

Ask the data

Answers come from the same store as the site; every number is checked against the record it cites.

Adaptation · Labour & consumers

Freelance writing earnings after ChatGPT (Upwork)

consistent with normal70/95 confidence, good evidence, some ambiguitygrade Bleadingrefreshed irregularly: A one-off platform study; refreshed only when a comparable study publishes.

1What this measures

Change in monthly earnings for writing-related freelancers on Upwork after ChatGPT's release relative to less exposed categories, from Hui, Reshef and Zhou, with the jobs effect alongside.

Why it matters. The most substitutable work on the most measurable platform. If displacement shows anywhere first, it is here, and it shows as single digits.

Proxy types
welfare, deployment
Unit
share
Cadence
annual
Valve
adoption to adaptation

2How we track this

Normal band
≥ -10.0% and ≤ 0.0%
Fast band
≤ -20.0%
Falsifying
—

Normal = declines under 10% in the most substitutable freelance categories, a tool effect at the margin; fast = a 20% or worse collapse in freelance earnings. Between is `emerging`.

Applied to cepr_voxeu.upwork_writing_freelancers.monthly_earnings_change.pt.

3Tracker interpretation

Five percent off earnings and two percent off jobs for the workers most exposed to a text generator. Real, small, early.

4Evidence

Latest point
-5.2%as of 2023-12-01
upwork_writing_freelancers
Value the bands apply to
-5.2%as of 2023-12-01

3 observations. Hollow points are disputed (see counterevidence). Every point links to its observation.

5Status and reasoning

consistent with normalsince 2026-09-10 · evaluate

Evaluator: -0.052 (CEPR VoxEU columns (via Wayback; cepr.org 403s bots), 2023-12-01) is inside the consistent band (lo=-0.1 hi=0). Auto-reason; band rationale: Normal = declines under 10% in the most substitutable freelance categories, a tool effect at the margin; fast = a 20% or worse collapse in freelance earnings. Between is `emerging`.

6Timeline notes

  • 2023-12-01 upwork_writing_freelancers · -5.2%as of 2023-12-01

7Counterevidence

What cuts against this reading

One platform; short window after release; the comparison categories were also exposed; later models may widen the effect.

8Update history

  1. 2026-09-10unmeasured to consistent with normalconf — → 70 · evaluate

    Evaluator: -0.052 (CEPR VoxEU columns (via Wayback; cepr.org 403s bots), 2023-12-01) is inside the consistent band (lo=-0.1 hi=0). Auto-reason; band rationale: Normal = declines under 10% in the most substitutable freelance categories, a tool effect at the margin; fast = a 20% or worse collapse in freelance earnings. Between is `emerging`.

    obs:55a1d3e6

9Confidence

70 / 95 — good evidence, some ambiguity

Confidence is independent of status: 90–95 multiple strong independent sources; 70–89 good evidence, some ambiguity; 50–69 mixed or hard to operationalise; below 50 limited or vague.

10Related