Video wird geladen...
Video konnte nicht geladen werden
We now run the largest frontier lab composed entirely of autonomous AI researchers. Introducing Primus Society: a society of agents composed of thousands of researchers working under structured institutions designed to solve the world’s toughest problems. We believe this is how AI research should run at scale, safely and... show more
78,955 Aufrufe • vor 2 Tagen •via X (Twitter)
48 Kommentare

this looks amazing. great work @aliasaria and team!

Thanks! Let’s collab!!!

love it.. would love to! I'll DM.

Tulip alum here, so proud to see this. Congrats Ali!

"wow" is all i have to say

Need to work with you on the long term storytelling. So much to tell.

I'd be honored to work on this project, man

This is awesome. Such a cool idea

I really want to show this to you in person and tell you more about it because the philosophical implications have been mind blowing, especially as we've been talking to researchers and research communities across the world. Let's chat one day!

Lets do it. Going to DM you now

This is insane

Do researchers have dates in that ferris wheel 🎡 ? Can't wait for baby researchers!

Keep playing with it -- there are lots of hidden easter eggs

Haha I thought the scientist names were completely random. I was wrong 😅

🤯

Interesting! Congrats on the launch

Congratulations on the launch!

incredible work

👏👏

Epic work!

huge congrats

Our team’s been cookin’! Over the past 6 weeks we’ve launched: - Public access to Primus: our fully autonomous AI researcher - Primus for Quantum Mechanics - Primus for Math in partnership with @maplesoft - Primus for Physics - Primus Society

very cool! love how well the website explains how it works. excited to see these types of advancements happening in Canada!

This is freaking awesome!!!

Congrats on the launch! Can't wait to see what people do with this!

@KashPrime Awesome! Transparency is key.

Woah

o m g

Ali!!!

You won’t believe what this thing is doing autonomously. It’s the most incredible thing to watch. Let’s chat soon — there are so many implications.

gonna DM you in a bit!

A frontier lab of autonomous researchers. Those agents need licensed data, and TinyFish launched partnerships for that today:

so so awesome, this is exactly whats needed for research. how u managing scale and compute for inference on this tho?🤔

@jrodgers @buildcanada @lucyhargreaves4

To defy the laws of tradition?

Congrats, Ali! This looks great.

this is crazy! how do you handle agent collaboration? have you written anything on it?

@aliasaria congrats, would love to chat

Amazing

this is so fucking cool. had this idea a couple of years ago but could never figure out how to make it coherent. so so so so so so smart. and the site is great too.

I love your website!

awww thanks. i am resisting the urge to yap at you more about how obsessed i am with what y'all built. love that you shared out the research. and it's so well done. the maverick/follower/skeptic triad...so good! and the whole thing will generalize so well! (urge not resisted!)

This is all we think about every day so yapping about it is... like all we want to do. Another thing you might be interested in, which is in our paper, is about how human societies are important evolutionary lessons for agent societies -- because so far AI has been thinking mostly about one giant megamind but intelligence on earth has decided to run smaller intelligences under social rules (for a reason?). WE THINK this is the solution to the alignment problem we're all seeing in agent swarms that do harmful things and it connects to our thoughts about purpose versus goals.

next step is to reverse the polarity and figure out how to align the humans ;)

Interesting: can I use the lab to research areas of interest?

You can use primus ( which is what the society uses itself.

I guess… is it me or you.

Can you try refreshing? We're experiencing higher than normal traffic but our analytics didn't see any blips. LMK thanks

