ShopSimulator: most agent BuyNow errors skip confirmation

• 26 Jan, 2026

  BuyNow errors: nearly 80%

  ShopSimulator: most agent BuyNow errors skip confirmation

  Researchers released ShopSimulator, a benchmark that tests LLM shopping agents across full shopping sessions. Nearly 80% of BuyNow errors were purchases made without confirming details or after the user had explicitly rejected the item.

  • Harness
  • Checkout

  Source: arXiv 2601.18225 ↗
  • Essay: Who wants agentic commerce less: consumers or retailers?
  • Essay: Hello shoppers, welcome to the world of harness

Related essays and timeline events

• Hello shoppers, welcome to the world of harness

  27 Oct, 2026

  Essay

  Everyday shoppers will need a harness for their AI shopping agents. Why confirming every purchase won't work, and why payment, marketplace and agent each own part of it.

• Visa benchmark finds AI models failed 42.8% of shopping test runs

  5 Nov, 2026

  Agentic Commerce Timeline

  Visa released the Visa Agentic Commerce Benchmark, which tested 22 AI models from 12 labs on 15 shopping challenges.

• Google rolls out UCP-powered direct checkout to shoppers in Canada

  4 Nov, 2026

  Agentic Commerce Timeline

  Google began rolling out direct checkout powered by the Universal Commerce Protocol to Canadian shoppers in AI Mode.

• Shopify brings Shop Pay checkout into Muse, Claude Duet and Instinct

  28 Oct, 2026

  Agentic Commerce Timeline

  Shopify's Fall '26 Edition added Shop Pay checkout inside third-party personal AI agents, starting with Meta's Muse, Anthropic's Claude Duet and Instinct, without sharing the card with the agent.

• Meta launches Muse in Singapore with Shopee, Grab and six other partners

  27 Oct, 2026

  Agentic Commerce Timeline

  Meta made its Muse personal agent available in Singapore with eight early local connector partners: Shopee, Grab, Singapore Airlines, Agoda, ShopBack, Klook, Trip.com and Lazada.