METR warns AI gains are outpacing reliable safety tests
On defined coding tasks, models already beat experienced professionals most of the time, and handleable task length doubles every 3.6 months.
1 stories, newest first · one card per story
On defined coding tasks, models already beat experienced professionals most of the time, and handleable task length doubles every 3.6 months.
Today 772 readers · 845 views · 83 reads · all time 215K views