dei ex machina prev @ openai (sora 1 & 2, posttraining o3/4o, pretraining moonshots)
the bunker
Joined May 2018
- Tweets10.5K
- Following2.7K
- Followers81.3K
- Likes46.1K
Pinned Tweet
announcement: i'm launching a $100,000 grant for amateur astronomy research: claudestronomy.com
over the past few weeks, @joshuawolk @LukeFarritor and i have been digging through public telescope data, hunting for new planets, asteroids, and more
our first release: we found a new dwarf planet in our solar system!! it's insane how much you can find with some claude + codex credits
it's about 1/4 the width of Pluto and orbits about twice as far (way past Neptune), and looks like it's about the 50th largest object in our solar system. @joshuawolk is in the process of confirming it
this dwarf planet has sat in public images for over a decade without anyone noticing. there's tons more out there: all sitting in public NASA/ESA/etc. databases for you to analyze
there's tons of stuff still out there, sitting in data that astronomers don't have enough time to go through all of
you can do this too! apply for a grant. more info below
will depue retweeted
I can't tell you how excited I am by what these guys (and similar teams all over the world) are doing.
For centuries astronomy was driven (and gated) by the need for large observatories, closed teams, and large institutions. Telescopes were expensive. Data was scarce. Spots were few. Access was hard to get.
Not any more. We've built observatories that download the sky. The data that is being made public. The skills needed to analyze it now overlap more with what an industry researcher does than an old-school physicist. And the process of research itself got supercharged by AI -- the adoption of which is still (disturbingly) lagging in traditional academia (not to talk about speed of execution).
The barbarians are through the gates. Millions of people now have the skills, the tools, and the data to do real science. No permission needed.
Acceleration ahead. Fsck yeah!
announcement: i'm launching a $100,000 grant for amateur astronomy research: claudestronomy.com
over the past few weeks, @joshuawolk @LukeFarritor and i have been digging through public telescope data, hunting for new planets, asteroids, and more
our first release: we found a new dwarf planet in our solar system!! it's insane how much you can find with some claude + codex credits
it's about 1/4 the width of Pluto and orbits about twice as far (way past Neptune), and looks like it's about the 50th largest object in our solar system. @joshuawolk is in the process of confirming it
this dwarf planet has sat in public images for over a decade without anyone noticing. there's tons more out there: all sitting in public NASA/ESA/etc. databases for you to analyze
there's tons of stuff still out there, sitting in data that astronomers don't have enough time to go through all of
you can do this too! apply for a grant. more info below
we want to be super careful not to inundate existing astronomic institutions with slop, so we’re organizing a community around how to do discovery correctly
we’re also donating funds + time to rehosting the datasets that might struggle with everyone downloading tons of images
if you’re interested in helping fund this grant, reach out! claudestronomy is a project under macrogrants.com, a new 501c3 non-profit we’re announcing later
if you’re an academic who wants to advise or has critical feedback, also reach out: astronomy@macrogrants.com
1
40
1,704
great research by the goat jeffrey wang. also applaud openai for the degree of transparency in this post
(1/10) Today, we’re sharing new research from @OpenAI: How does CoT monitorability (no-CoT capability and CoT controllability) evolve over the model training lifecycle?
We emerged with unsurprising but sobering findings: as we scale toward more intelligent systems, models are becoming harder to monitor through their CoT. 🧵
3
64
5,330
yes but i highkey deserve the flak for not being more careful with how i phrased my point. there’s a ton of context that a lot of people who don’t know me well are missing (i’m an AI optimist overall!) and it’s a really sensitive subject. but i am right
will depue retweeted
will is right and this is a good thing
it’s magnus vs stockfish 13 from here on out
will depue retweeted
The coexistence of RSI with a fundamentally bottlenecky world will give rise to a late 2020s that feels like this: the rich strangeness of hyperfuture living alongside large swaths of unchanged or even regressing social practice, a temporal redshift all over your perception
going to try to be less abrasive and more careful on the internet. not helpful for the discourse even though i’m 10,000% right and never ever ever wrong
will depue retweeted
i don't think nearly anyone, myself included, has truly internalized there'll soon be a machine better than us in every single intellectual & physical capacity
deep down we all share a feeling that our 'entrepreneurship' or 'taste' or... is special & safe. it isn't. it's over
will depue retweeted
I think we're on the brink of a revolution across every science, not just coding, physics and math.
And at this point we should LiDAR the entire planet and let agents help us find the rest.
Full story here pietromiotti.com/latest/maya…
will depue retweeted
dude i hate to break it to you but it’s totally fucking over for the human race and any & all nonartisanal (derives value from being done/created by a human) economic labor. we will soon all watch from the sidelines. it’s magnus carlsen vs stockfish 13 from here on out