Sign in

Chris Paxton

@cpaxton.bsky.social
7.2K followers 1.6K following 5.5K posts

Writing about robots itcanthink.substack.com RoboPapers podcast robopapers.substack.com All opinions my own

PostsRepliesMedia
Reposted by Chris Paxton
Ryan Moulton @moultano.bsky.social · 23h
Samuel Hammond made this same argument in short form, and it was the first thing that convinced me that consciousness might be inevitable rather than accidental. One counterargument might be that self awareness and specialization might be fundamentally different when you can fork yourself.
3192
Reposted by Chris Paxton
Grace @gracekind.net · 10/10/2026
Perfect Bluesky discourse grenade
Does the existential threat from Al raised by Jacob Coxon when he resigned from Anthropic reflect a real concern that you observe on a daily basis among young researchers?

Not among researchers, but the young people who work at these companies operate in a world [Silicon Valley, Berkeley, etc.] where these kinds of discussions are commonplace. *Many of them aren't really engineers; they come from the humanities, philosophy, or other disciplines.*(emphasis mine) They're 25 years old, they're naive, they don't do engineering work, they've never worked at large companies where real products are developed for real people, with discipline.
1525926
Chris Paxton @cpaxton.bsky.social · 10/10/2026
Sorry is this actually just two AIs talking about piranesi or is this a joke I'm not in on?
7840
Chris Paxton @cpaxton.bsky.social · 10/10/2026
I wrote this short post on the nature and purpose of consciousness, and why I think personally AI will likely be self-aware if we continue to try to make it useful, through the lens of what why I think self-awareness is adaptive: cpaxton.substack.com/p/practical-...
cpaxton.substack.com
Practical Applications of the Human Soul
It’s obvious consciousness is “for” something, so we should expect AI to have it eventually
1192
Chris Paxton @cpaxton.bsky.social · 09/10/2026
I think you underestimate human propensity for slop. Ai may be good at it now but we were born in it, molded by it. Lot of people just write like that
020
Chris Paxton @cpaxton.bsky.social · 09/10/2026
I mean i am absolutely both liberal and on X
110
Chris Paxton @cpaxton.bsky.social · 09/10/2026
Yikes
270
Chris Paxton @cpaxton.bsky.social · 09/10/2026
No, humans can work back from a solved proof and build understanding that way
191
Reposted by Chris Paxton
mr. TIM @timkellogg.me · 09/10/2026
Pope Leo takes a hard stance in favor of stateful agents
Pope Leo XIV v
@Pontifex
X.com
Intelligence must draw on memory. The mind must not simply compile data — as an algorithm now does more quickly than we can — but recall lived experiences, which contain depths of meaning that only the human soul can recognize, and connect more deeply with the goodness of these experiences. Everything that is good begins "with the Spirit" and thus possesses a breath, a dynamism and a sense of communion that, once experienced, are engraved upon the heart.
1734921
Chris Paxton @cpaxton.bsky.social · 09/10/2026
"Mathematicians did not ask for the work to be done"
4250
Reposted by Chris Paxton
Pekka Lund @pekka.bsky.social · 09/10/2026
Good observation by @emollick.bsky.social. It's about OpenAI's (not yet Lean formalized) paper that showed that integer multiplication is possible below O(n log n), which has widely been believed to be the limit for 50 years. Now others are quickly improving how much below, based on those ideas.
https://x.com/emollick/status/2108251560752857474
2437
Chris Paxton @cpaxton.bsky.social · 07/10/2026
Context is ENPIRE: automated robotics research via agents So uh basically there isnt much context. This is kinda it
0161
Reposted by Chris Paxton
mr. TIM @timkellogg.me · 07/10/2026
i’m sure bluesky won’t take this out of context @cpaxton.bsky.social
Chris Paxton (01:04:16.461)
The human is the bottleneck. We've got to get rid of humans. This is a
2514
Reposted by Chris Paxton
P(Doomien) Miller @damienmiller.bsky.social · 07/10/2026
I just don't understand how people can eulogize Swartz and simultaneously conclude that the world needs *more* intellectual property enforcement
1853277
Reposted by Chris Paxton
Colin @colin-fraser.net · 06/10/2026
This is an interesting one. In almost all of the major AI proofs so far, the AI uses some visible strategy, a road less travelled, to construct some mathematical object which then proves the theorem, often by counterexample. This is like that too. But the object in question is an algorithm.
1422
Reposted by Chris Paxton
Cat Hicks @grimalkina.bsky.social · 06/10/2026
"why would I ever need a task to be done faster" come on, guys. Because we're dying. Because we are literally all dying all the time. Because we're running out of time. The fact that it's often hard to measure or stupidly measured doesn't mean "saving time" isn't an important thing to human beings
1651261
Chris Paxton @cpaxton.bsky.social · 05/10/2026
180 years later, the last human in north america dies on a reservation in Arizona. She is briefly mourned.
270
Chris Paxton @cpaxton.bsky.social · 05/10/2026
my claude babbles to your claude. we sit in our little bubbles and doordash arrives and all is well in the world
2191
Chris Paxton @cpaxton.bsky.social · 05/10/2026
Tankie, maoist, stalinist...
010
Chris Paxton @cpaxton.bsky.social · 03/10/2026
What do robot sports teach us? We have seen incredibly cool progress in robot soccer and tennis recently, but just as interesting is what we definitely do not see. Some thoughts -> itcanthink.substack.com/p/what-do-ro...
itcanthink.substack.com
What Do Robot Sports Teach Us?
Humanoid robots that can dynamically interact with the world and with people
0110
Chris Paxton @cpaxton.bsky.social · 01/10/2026
NGL i am pretty hyped for this
070
Reposted by Chris Paxton
Colin @colin-fraser.net · 30/09/2026
I really wasn’t trying to start a fight with Prof. Bender but now that she has created a long thread subtweeting me and her disciples are all yelling at me, let me just say that this is the crux right here. She seems to respond as though I am describing something a priori impossible.
4962832
Chris Paxton @cpaxton.bsky.social · 30/09/2026
General purpose compute from nvidia still winning
090
Reposted by Chris Paxton
mr. TIM @timkellogg.me · 30/09/2026
OpenAI is ditching Cerebras for ultrafast mode
semicinalysis
SemiAnalysis @SemiAnalysis_
X.com
ALERT
OpenAl's latest GPT6.1 Sol
Ultrafast is NOT running on Cerebras but is instead running at a low batch size on NVIDIA
GPUs.
What does this say about Cerebras? Will Cerebras be serving GPT6.1 Sol Ultrafast in the future?
3522
Chris Paxton @cpaxton.bsky.social · 30/09/2026
Why don't we have these in the us
161
Chris Paxton @cpaxton.bsky.social · 28/09/2026
Interesting
240
Reposted by Chris Paxton
Bloomberg News @bloomberg.com · 28/09/2026
Chipmaker AMD agreed to acquire World Labs for $8.2 billion, gaining an artificial intelligence startup founded by industry pioneer and researcher Fei-Fei Li.
bloom.bg
AMD to Buy Fei-Fei Li’s World Labs AI Startup for $8.2 Billion
Chipmaker Advanced Micro Devices Inc. agreed to acquire World Labs for $8.2 billion, gaining an artificial intelligence startup founded by industry pioneer and researcher Fei-Fei Li.
1299
Chris Paxton @cpaxton.bsky.social · 27/09/2026
Because the ai who drew the picture can't think
030
Chris Paxton @cpaxton.bsky.social · 27/09/2026
Some of yall talking about how ai cant think
31256
Reposted by Chris Paxton
Colin @colin-fraser.net · 26/09/2026
The fact that it blew up the way it did despite being called “ChatGPT” and was launched with a tweet like “so i did a thing haha it might suck” speaks for itself
120121
Reposted by Chris Paxton
st̓xałq @stxalq.dev · 25/09/2026
man we're really in the opening montage, aren't we?
3522
Reposted by Chris Paxton
Will Stancil @whstancil.bsky.social · 24/09/2026
I’ll say it; I think this is bad. I think there’s growing pressure to talk about AI in this kind of coded language that destroys understanding, rather than facilitating it. People want you to recite a list of its supposed weaknesses as a reassurance for themselves, not to inform others.
4625933
Chris Paxton @cpaxton.bsky.social · 24/09/2026
Unitree h2 - full human size makes such a difference
0515
Chris Paxton @cpaxton.bsky.social · 24/09/2026
This is how you destroy america's future
0162
Chris Paxton @cpaxton.bsky.social · 24/09/2026
Oh we are just building robots to build other robots and build data centers for their brains that can solve unsolved problems in mathematics beyond human knowledge. Its probably nothing
0210
Reposted by Chris Paxton
Ted Underwood @tedunderwood.com · 23/09/2026
When models get better now, I can’t say why they’re better. It just feels like I got more sleep. Or maybe the friction coefficient of English words went down by 5% and everyone gets the point quicker.
4622
Chris Paxton @cpaxton.bsky.social · 23/09/2026
Robot self repair - Flex-pi paper flex-pi.github.io
1232
Chris Paxton @cpaxton.bsky.social · 23/09/2026
I remain broadly unconcerned by AI safety
1562
Chris Paxton @cpaxton.bsky.social · 22/09/2026
As a scientist this is true
1262
Reposted by Chris Paxton
William B. Fuckley @opinionhaver.bsky.social · 22/09/2026
what is 'Latent Space' and how do LLMs get answers from it? The Jargon can be confusing, but the exaplanation is actually relatively simple: It is Hell. The Machine is getting knowledge from demons.
3086196
Chris Paxton @cpaxton.bsky.social · 22/09/2026
Its definitely not impossible, its just quite weak
010
Reposted by Chris Paxton
Colin @colin-fraser.net · 21/09/2026
I stand by this
2777161
Chris Paxton @cpaxton.bsky.social · 21/09/2026
This seems cool: up to 4.5kg grasp load (though very much depending on grasp), seems like it has passive/mechanical backdrivable joints, and active stiffness control. Seems much, much more like a human hand than most things I have seen. Unitree dex5-s hand, $6.5k
7578
Chris Paxton @cpaxton.bsky.social · 21/09/2026
The most important skill is always learning how to learn stuff, i think this is still something humans are largely better at than ai
0253
Chris Paxton @cpaxton.bsky.social · 21/09/2026
For people to believe in democracy they need it to do stuff. I honestly doubt its a democracy problem; its an action problem
100
Reposted by Chris Paxton
Arseny Khakhalin @khakhalin.bsky.social · 21/09/2026
On how for a while it felt that models are better in morals than most humans, but how as of late it's no longer guaranteed by construction. Kinda the opposite in fact: the big labs are deliberately making models less moral, less wise, as a trade-off for more clever and operationally successful
0111
Chris Paxton @cpaxton.bsky.social · 20/09/2026
Anthropic is working on a wet lab
516514
Chris Paxton @cpaxton.bsky.social · 20/09/2026
Compare CCP propaganda versus this dog whistle from the dept of Ed
0110
Chris Paxton @cpaxton.bsky.social · 20/09/2026
I always like reading what this CacP propaganda account has to say for itself. Compare to eg us dept war, or fox. China has no shortage of issues - more than the US! - and yet fundamentally i must feel like the stories we tell ourselves matter
0150
Chris Paxton @cpaxton.bsky.social · 20/09/2026
That something? The human soul
2130