Bluesky Thread

i woke up still thinking about Dario’s take

January 22, 2025 View original thread

i woke up still thinking about Dario’s take

character > capabilities

it honestly seems like anthropic’s moat. it’s quite an astonishing thought that reasoning is *just* a trait, not worthy of defining an entire model on

Tim Kellogg @timkellogg.me

Dario hints that "reasoning" is just a capability, and the dichotomy between "reasoning models" and "regular models" is void

exactly what i've thought — sonnet, phi4, there's a lot of these models that think first. and r1 annoys me that it feels **forced** to think

youtu.be/snkOMOjiVOk?...

21 2

might as well tack this on. Amanda Askell being the character lead on Claude. if you haven’t followed her work you should. she has an utterly fascinating way of viewing LLMs

bsky.app/profile/timk...

Tim Kellogg @timkellogg.me

hahahaha yes

The tweet reads:

“the only moat left in AI is amanda askell”

Posted by Minh Nhat Nguyen (@menhguin) at 12:06 PM on 1/20/25 with 1.8K views.

Dario’s take on revoking the AI exec order gave me a chuckle. basically, “it’s a bad move but the EO didn’t do much anyway so it’s not really a loss either”

it’s apparent in his demeanor that he’s working hard to stay out of Trump’s crosshairs while also maintaining his code of ethics

which is more than what can be said about mr altman

More like this