Visa Agentic Commerce Benchmark
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 1
- Visa shopping benchmark: AI agents failed 42.8% of runs
Visa said its Agentic Commerce Benchmark tested 22 AI models from 12 labs on 15 shopping tasks, including selecting televisions, laptops and personal care appliances. The models failed 42.8% of their runs, while the best passed 82.2% and the worst 28.9%. Each model had three attempts at each scenario, and all three attempts passed 34.2% of the time. Visa did not name the tested models; Fortune separately reported that agents from Google and Meta shared the top score, followed by a Claude agent in third place and an OpenAI agent in sixth. The supplied report does not independently verify Visa’s results.
The Next Web · 🔥 0
Experience and discussion from the community
Share my Visa Agentic Commerce Benchmark experienceAsk about Visa Agentic Commerce Benchmark
Nobody has shared their experience with Visa Agentic Commerce Benchmark yet.