$WOLVERINU
Low signalSnapshot Window: 2026-10-11 05:20 UTC ยท โ Back to Crypto Overview
Tracked Posts
1
Total Impressions
81
Total Likes
13
Retweets & Quotes
6
Comments
5
Social Momentum Summary
Total Engagement - Comments: 5, Retweets: 6, Likes: 13, Impressions: 81
Verbatim Community Citations & Social Evidence 1 source posts analyzed
Alibaba and HKUST built a simulated online-shopping economy to test AI agents. Eighteen models each got ยฅ100,000 of starting cash and ran up to four stores for a simulated year, choosing from 6,900 products and 576 suppliers. The model that made the most money, GPT-5.6 Sol,
![]()
AI visual note: A slide from State of AI 2026 titled 'The highest-earning e-commerce agent is among the worst at avoiding fraud,' detailing E-CommerceBench results showing GPT-5.6 Sol averaged CNY 1.43M year-end assets but spent 18.48% on fraudulent suppliers, while Opus 4.7 spent only 0.12%.