Stanford postdoc working on morality, happiness, and other subjective experiences that are really hard to measure.
San Francisco, CA
- Tweets 3,704
- Following 1,447
- Followers 2,642
- Likes 136,087
Pinned Tweet
Hi friends! I'm finally feeling well enough to share some news that I've been very excited about: earlier this week, my kidney was removed through a small cut in my stomach, tucked into a box, flown down to johns hopkins, and implanted into a stranger where its working great
91
238
6,545
Vlad Chituc retweeted
i find this especially amusing because i'm teaching ELIZA in Minds and Machines (Stanford's intro cogci w/ @mcxfrank) tomorrow...
People often eval user models on the "User Sim Index" - we probably shouldn't. I attached code for a "user model" scoring 95% on its 4 behavioral dims: 97% comms, 97% info, 92% clarify, 95% error reaction. I think this beats any released model. See also a sampled conversation
1
3
34
5,388
And just like that, it's Amazon Prime Day.
Enjoy: wired.com/story/absolute-bes…
36
184
1,751
271,174
Beliefs should pay rent! Unless it has to do with a topic you really, really, really believe in and think is important. Then you can just keep saying "but it's not empirically detectable yet" for as long as you want. Rationality, baby!!
IF we were to see catastrophic dynamics from AI, it wouldn’t happen overnight. We’d get upticks and signs. Unemployment won’t jump from 4% to 50% in a year.
Hence lack of big moves in unemployment yet *IS* data. The post has many examples.
I refer y’all to Yudkowski circa 2007.
1
2
119
Vlad Chituc retweeted
The most important briefings of the day.
Karl Deisseroth, professor of bioengineering and of psychiatry and behavioral sciences, shares the news of his Nobel Prize with his kids.
524
4,631
36,482
4,503,078
Vlad Chituc retweeted
I'm recruiting a grad student at Florida this fall! If you find boredom interesting, and psychological richness weirdly compelling, consider applying! Pumpkin carving & rock painting skills a bonus.
Applications due 12/1, see website: erinwestgate.com/join-the-la…
1
1
7
438
Vlad Chituc retweeted
I am still just SO excited that Felipe De Brigard got a MacArthur Genius Grant!
He has done so much amazing work in experimental philosophy, cognitive neuroscience, and a million other fields, and it's absolutely wonderful to see him getting recognized like this
We're thrilled to share that Felipe De Brigard has been named a 2026 MacArthur Genius Fellow for his work reshaping our understanding of the nature and function of memory. Felipe is the co-editor of "Neuroscience and Philosophy II,” publishing next month: bit.ly/4ABPYuF
ALT A photo of Felipe De Brigard next to text that reads "2026 MacArthur Genius Fellow"
1
3
50
1,954
Vlad Chituc retweeted
your homunculus will never be ensouled bro, stop showing the pope your homonculus, he doesn't like it
NEW: According to a bombshell report in the New York Times, Anthropic co-founder Chris Olah threatened to walk out of Pope Leo XIV’s AI encyclical launch in May because the pope rejected the idea that machines can be conscious.
Olah’s team then privately lobbied the pope’s advisers “to take the possibility of model consciousness seriously.” Pope Leo XIV held firm.
For months, Anthropic has wined and dined theologians and religious scholars under nondisclosure agreements, hoping they would bless the idea that Claude has moral standing. thelettersfromleo.com/p/nyt-…
13
895
9,018
142,712
Vlad Chituc retweeted
Pouring one out for MTurk today
ALT Amazon Mechanical Turk will permanently close on September 30, 2026. For Workers and Requesters currently using the service, please find FAQs below that explain how you can prepare for this closure:
2
15
806
Vlad Chituc retweeted
🚨Come work with us! 🚨
We're hiring a postdoctoral fellow for Stanford's AI for Public Benefit Lab (ai4pb.stanford.edu), in partnership with @StanfordHAI
Open to PhDs from psych, poli sci, sociology, CS, econ, HCI, data science, and related fields. Please share widely!
Position + how to apply: postdocs.stanford.edu/prospe…
2
40
181
14,867
Vlad Chituc retweeted
looks like you pretty much covered it
How do we explain to younger generations that short sleeve tshirt over a long sleeve used to be so popular they used to make fakes which were just long sleeve tees with different sleeves stitched in to look like it's layered
39
1,351
55,902
1,068,816
Vlad Chituc retweeted
socrates v.s. didion would have been in the top 10 anime battles of all time
1
56
1,176
22,389
People with more vivid mental imagery show pupil dilation responses and people without it don’t lol pmc.ncbi.nlm.nih.gov/article…
when you imagine the sun do you have to close your eyes or do your pupils dilate because of the light? NO. its semantics and always will be
1
4
717
The funniest thing about the aphantasia debate is that people’s inability to imagine minds that work differently than theirs means you regularly see comments exactly like this from both sides saying the other side is ACTUALLY like them and just confused about semantics
I'm like 99% certain that aphantasia isn't a real thing. It's a semantics issue. People are seeing that "can you see the apple?" chart and are being led to believe other people can project a hallucination of an apple into their physical vision through sheer force of will.
1
1
15
1,631
Vlad Chituc retweeted
Some worry that AI-risk assessment is basically coming from a (rationalist) monoculture. And so for instance aggregating p(doom) estimates is less informative than you’d like, because the estimates aren’t independent. A possible check on that is to track Chinese views.
“My friends in Silicon Valley always ask me what the AGI timeline is for Chinese policymakers, and what the key doom scenario is for Chinese policymakers… They assume Chinese officials think exactly like Silicon Valley executives.”
Really great interview:thewirechina.com/2026/09/24/…
1
6
261
The amount of people who think we can’t do magic because we’re too dumb—and not that magic isn’t real—is genuinely baffling to me. There are so many trivial things (like the knapsack problem) that are computationally intractable, why do we think psychohistory is plausible?!
The logical endpoint of AI development in the social sciences is psychohistory from Foundation.
A predictive world model which can simulate with high accurate future events and counterfactuals
5
431
Vlad Chituc retweeted
Capitalizing on the ~Western Mass~ discourse to say I love it here and am still hiring a post-doc!
We plan to hire a post doc to start in Fall 2026 (flexible), in the Aulet Lab at UMass Amherst, funded for two years. The postdoc will help establish our developmental fMRI research program studying symbolic number acquisition in 4-8 year old children (among other things!).
2
3
238
This is something I've found really weird about the discourse on AI risk, where people seem to habitually talk about AI as "other kind of agent and kind of being" and not "tool humans use." We don't need to charge ChatGPT — like just apply tort law. That's why we have tort law.
All anyone needs to do as far as AI safety goes is bring charges against the labs for hacking other companies and the liability exposure will then make them invest in security.
2
1
5
755