Megan Kinniment
@MKinniment
I like agents, human or otherwise. @METR_Evals
Happy for this to be released! It’s the result of a lot of hard work from many of us at METR :) A big question is whether these results apply to ‘real’ tasks. Here’s some thoughts on that:
When will AI systems be able to carry out long projects independently? In new research, we find a kind of “Moore’s Law for AI agents”: the length of tasks that AIs can do is doubling about every 7 months.
When I ask the models for help, but they can’t do it either:
AI-enabled coups seem like an important (and understudied) topic. Happy to see this work, and especially all the suggested interventions!
New paper on AI-enabled coups. When AI gets smarter than humans, a few leaders could direct insane amounts of cognitive labor towards seizing power. In the extreme, an autonomous AI military could be made secretly (or not so secretly!) loyal to one person. What can be done? 🧵
Very cool to see METR’s work in the Financial Times. AIs have different capability profiles to humans, so we might be surprised by the order in which different tasks are automated. Great piece by @jburnmurdoch.
There has been, to date, little evidence of AI causing large-scale disruption to the labour market, even in occupations that are reportedly at very high risk. What’s going on? on.ft.com/4l8boHI