INDUSTRY NEWSInvestigateNEXT 12 MONTHS
Visa Finds AI Models Fail 43% of Shopping Test Runs
PYMNTS AI
Factual evidence
What the source reports
Visa evaluated 22 AI models across 12 labs, finding a 42.8% failure rate when executing complex shopping scenarios.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- PYMNTS AI · INDUSTRY NEWS
- Published by source
- 5 November 2026
- Collected by OneBench
- 6 Nov 2026, 08:01 UK
- Original headline
- Visa Finds AI Models Fail 43% of Shopping Test Runs ↗
Stored source excerpt
Visa found that when 22 artificial intelligence models from 12 labs were put through realistic shopping scenarios, 57.2% of the test runs passed and 42.8%…
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 transaction failure rates in current AI models highlight reliability limits for autonomous agentic commerce.
Do what
Review model evaluation standards and reliability metrics with the team responsible for payment product strategy.