CMU's Odysseys2 Benchmark Exposes Web Agents Failing 53 of 100 Real Tasks

A Carnegie Mellon and Browser Use benchmark of 100 hour-long live-web tasks exposes how far web agents still are from reliable knowledge work.

Read the full story at AlphaSignal ↗

Timeline · 1 report

2026-10-28 04:42 · AlphaSignal
CMU's Odysseys2 Benchmark Exposes Web Agents Failing 53 of 100 Real Tasks

More stories

Gemini 4 Carbon: our next frontier model — Google Gemini Blog
Anthropic reveals plans for blockbuster IPO — Sky News Technology
OpenAI and ServiceNow expand partnership across enterprise workflows — OpenAI News
OpenAI launching election war room to track interference efforts — Politico Technology
Google's Gemini 4 Carbon Tops AI Rankings and Hunts Security Bugs for Days — AlphaSignal
CoreWeave Builds the First Global Backbone Network Purpose-Built for AI — CoreWeave Blog
Microsoft recorded $10 billion in revenue from OpenAI — TechCrunch AI
PyTorch 2.15 Release Blog — PyTorch Blog

Get the daily brief of stories like this at 6:30 every morning →