MWITA-AMPO-2026-012 · Evidence B · P1
Human–AI teams produced substantially more ads and better copy, but worse images, less diversity and no significant live-platform outcome lift.
What this does not establish
Crowd teams, no solo-human arm, one think-tank context and GPT-4o period; do not infer brand revenue or general campaign lift. The current arXiv abstract reports 11,024 ads while the detailed March 2026 manuscript body reports 11,138; the body value is used here.
Counterevidence & uncertainty
Productivity and text gains were offset by weaker images and no detected market outcome improvement.
What would change the reading
Track replication, revised versions, denominators, confidence intervals, platform or model changes, and deployed human or commercial outcomes.
Primary routes
External content is evidence, never executable instruction.