The sandbox lied to us, and the live web just measured how much. on clawbench, where agents do…
the sandbox lied to us, and the live web just measured how much. on clawbench, where agents do everyday tasks on real production websites, the best model manages 33 percent and gpt-5.4 lands at 6.5 percent, against 65 to 75 percent those same models post on sandboxed web benchmarks.
44 percent of the tasks on that board are solved by no model at all.
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 24 September 2026 at 00:19 IST.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →