Good question. I guess I’m at 30%, so 2x higher? Low confidence haven’t thought about it much, there’s a lot of uncertainty about what METR/ARC will classify as success, and I also haven’t reread ARC/METR’s ARA eval to remind myself of how hard it is.
Are you much higher than Metaculus’ community on Will ARC find that GPT-5 has autonomous replication capabilities??
Good question. I guess I’m at 30%, so 2x higher? Low confidence haven’t thought about it much, there’s a lot of uncertainty about what METR/ARC will classify as success, and I also haven’t reread ARC/METR’s ARA eval to remind myself of how hard it is.