Kabir Kumar, aiplans.org
banner
kabirkumar.bsky.social
Kabir Kumar, aiplans.org
@kabirkumar.bsky.social
I run AI Plans, an AI Safety lab focused on solving AI Alignment before 2029.
For several weeks I used a stone for a pillow.
I once spent a quarter of my paycheck on cheese.
Ping me! DMs not working atm due to totalitarian UK law :(
SurpassAI
Pinned
Working on a meta plan for finding an alignment solution, I'd appreciate feedback & criticism please - the more precise the better.
I'm calling it - Woke is back
December 2, 2025 at 3:24 PM
and so the curse spreads
What’s the lore behind choosing your career path ?
November 29, 2025 at 2:33 PM
there needs to be a new benchmark to just shame companies into not doing this anymore
November 27, 2025 at 5:42 PM
www.reuters.com
November 21, 2025 at 9:51 PM
Openai, more like openbubble
November 21, 2025 at 11:22 AM
I don't understand why Spotify doesnt just do a CostCo like model, or hell, even a Youtube like model, where people can pay for premium and then on top of that buy an album or some other exclusive on the platform.
youtu.be/gLv0nEvpOEc?...
Why Spotify feels so different now
YouTube video by Morning Brew
youtu.be
November 18, 2025 at 2:25 PM
Reposted by Kabir Kumar, aiplans.org
this is literally never not accurate
November 10, 2025 at 6:32 PM
This is a collab with DeepMind & Terence Tao btw
November 6, 2025 at 10:32 AM
the deepmind team is so cracked its crazy
November 5, 2025 at 5:56 PM
What's the ml equivalent of this?
What other seemingly obligatory phrases do you notice in neuro papers?
November 5, 2025 at 3:50 PM
Holy shit
November 5, 2025 at 9:29 AM
I hate EA haloween, what do you mean you're The Antichris
November 2, 2025 at 6:11 PM
The wikipedia article for Existential risk from artificial intelligence never mentions Yudkowsky in the actual text, only in the sources. It's generally quite low quality.
The Talk suggests that it's being done by people who largely don't know much about this topic
en.wikipedia.org/wiki/Talk:Ex...
Talk:Existential risk from artificial intelligence - Wikipedia
en.wikipedia.org
October 27, 2025 at 3:04 PM
Opinions on lettuce bun burgers?
October 24, 2025 at 4:50 PM
i cancelled my chatgpt subscription
October 24, 2025 at 6:02 AM
October 22, 2025 at 1:36 PM
jesus christ, wtf are they doing to gemini.
October 22, 2025 at 11:39 AM
in ai safety, unfortunately the appearances of the jesters and oracles are swapped
October 21, 2025 at 7:07 PM
'if we regulate AI, China will win!' Meanwhile, China:
October 21, 2025 at 9:40 AM
MoneyMan!! He has the Power of Money!!
October 17, 2025 at 2:31 PM
oh damn
Very excited to be able to talk about something I've been working on for a while now - we're working with Commonwealth Fusion Systems, IMO the leading fusion startup in the world, to take our work on AI and tokamaks and make it work at the frontier of fusion energy. deepmind.google/discover/blo...
Google DeepMind is bringing AI to the next generation of fusion energy
We’re announcing our research partnership with Commonwealth Fusion Systems (CFS) to bring clean, safe, limitless fusion energy closer to reality with our advanced AI systems. This partnership...
deepmind.google
October 17, 2025 at 2:32 AM
Prestige Maxing is Killing the AI Safety Field
October 16, 2025 at 4:36 PM
Very much agreed!!
Welcome to bluesky, make it better by using the For You feed: bsky.app/profile/spac...
October 16, 2025 at 1:42 AM
Guess who's joined the team as an alignment research advisor :)
www.lesswrong.com/posts/YJRb6w...
www.lesswrong.com
October 16, 2025 at 1:42 AM
Im travelling very late at night and finding that for a lot of niche things, like a place open late at night nearby where I can stay and wait for my train when the station open, GPT5 Thinking is not as good as just asking a person nearby who seems to work in the area and has a high vis vest.
October 16, 2025 at 12:44 AM