The Big OnePayments & Fintech
Visa's AI shopping benchmark reported 42.8% of test runs did not pass
In Visa's benchmark, 57.2% of test runs passed when 22 AI models from 12 labs went through realistic shopping scenarios. The best-performing models passed 82.2% of runs, while the worst-performing passed 28.9%, according to the release.
NoteFor payments firms, the gap between the best and worst models is the detail to watch. The benchmark is designed to create a roadmap for industry players, and the release says consumers can also use the test.
Read the original
The Morning Note, by email
The AI-in-finance stories that matter, before work
One short email each morning, readable in under three minutes. Free, and you can leave anytime.