Day 3 of 4
I Sold AI the Same Sweater Three Different Ways.
I gave an AI shopping agent different versions of the same cardigan to see whether the way a product is described changes what it recommends.
Follow the experimentAI’s weekly manager briefs averaged Across recall, accuracy, and relevance over four weekly briefs. across recall, accuracy, and relevance.
By April Gittens From I Let AI Prepare Me for Every Meeting With My Manager
Four-week experiment scorecard 20 daily briefs · 4 weekly briefs
Hypothesis supported 100%
Weekly briefs were perfect on quality. Recall, accuracy, and relevance all reached 100%.
Recall
97%
100%
Accuracy
91%
100%
Relevance
93%
100%
Average per brief Daily Weekly
AI credits 22.665 18
Tokens ~325.3K ~227.2K
Cost $0.227 $0.180
Weekly synthesis used 30% fewer tokens See the methodology
More experiments
Week 1 of 8
I Let AI Tell Me What My Body Needs.
Testing whether an AI-powered wearable can understand my biometric data well enough to give advice I’d actually follow.
Follow the experimentFall 2026
I Gave an AI Agent My Debit Card for a Week. Here's What It Bought.
Measuring whether an AI agent can autonomously research, select, and purchase products with real money while operating within predefined spending constraints.
Preview the hypothesis