Task-Completion Time Horizons of Frontier AI Models¶
Author or institution: METR
Kind of work: empirical AI capability evaluation
Open the work or publisher record
Access may require a library or purchase. The linked edition retains its own rights.
Why read it?¶
Agent task horizon measurement and capability growth; benchmark-specific, not a general AGI forecast.
Reading scope¶
No passage or chapter assignment is recorded. Consult the associated study route.
Study connections¶
No subject-level reading route is attached. This record may support a specific outcome or a broader cultural encounter.