INDUSTRY NEWSMonitorNEXT 12 MONTHS
Visa benchmark finds AI models fail 43% of shopping test runs
Finextra AI
Factual evidence
What the source reports
Visa benchmarked leading AI models on realistic agentic commerce tasks, reporting a 43% failure rate across test runs.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- Finextra AI · INDUSTRY NEWS
- Published by source
- 9 November 2026
- Collected by OneBench
- 10 Nov 2026, 08:01 UK
- Original headline
- Visa benchmark finds AI models fail 43% of shopping test runs ↗
Stored source excerpt
With agentic commerce beginning to gain traction, Visa has put a slew of models through their paces by giving them realistic shopping scenarios. The models…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.
OneBench interpretation
Institutional assessment
So what
High failure rates in agentic shopping models signal premature reliability risks for automated retail payment execution.
Do what
Review transaction-authorization limits and user-confirmation checkpoints with the team responsible for payment operations technology.