my notes on whatever i find interesting ✨ 🔬 science-backed 🔍 curiosity driven 🎤 host on @SciShow @KQED @PBSTerra @FeministBookClub 📍 san francisco/london
San Francisco
Joined June 2017
- Tweets1.2K
- Following413
- Followers1.2K
- Likes5.3K
Pinned Tweet
Your brain deserves better than brain rot 🧠
Come learn with Big Ideas: compound eyes 🐝 eggs 🥚 bioluminescence ✨
Human made. Real scientists. Real filming.
Out now: youtube.com/watch?v=5UCQoFO9…
from the @kqedscience team behind @deeplook
Is AI helping you think, or helping you spiral? 🌀 You ask → AI answers → You read and ask again → Repeat.
But what if AI knew when to stop?
What do our thought spirals and persistent AI agents have in common? (more than you might think)
📖 8.5 min: buddhismai.substack.com/p/cl…
Niba 🎬 retweeted
I'm starting to hate AI video essay scripts even more than AI visuals. When someone uses AI images or clips in a youtube video, it looks tacky for sure. But I can sympathize with having something you want to say, and getting stuck trying to figure out what to show on screen as you say it. I write myself into a corner all the time when scripting my videos. But when the foundation that a video essay is built on is itself created by AI, why even make the video in the first place? What creative vision did you have?
someone emails you and you notice it typo in their email where they forgot to put a space between two letters do you…
67%ignore it
0%point it out to them
33%complain on X
6 votes • Final results“nothing is set in stone, we rly want your feedback!”
the stone:
🪨 ur entire responsiblities already mapped out in a doc
I think good collaboration is mostly just making it easy for people to tell you what they actually think
good process is so underrated. Not that process is exciting, but bc when it works, u get to spend way more time making things and way less time figuring out what’s happening
Robots can move fast, but can they think fast enough to keep up?
⚡A new Al technique lets robots plan their next move while they're STILL acting, cutting reaction delays by 10x+ and completing some tasks 2x as fast, w NO extra compute.
My latest: livescience.com/technology/r…
nothing says “we’d love your input” quite like opening a document and discovering your future responsibilities have been filled in for you
;_; I'm so jaded now the first thing I wonder when I see unbelievable animal cuteness like this is whether or not it's AI
hot take that shouldnt be a hot take: it’s fair for jobs to ask for work samples or tests. But for a 9-12wk fellowship, the amount of unpaid labor just to apply can get pretty absurd. At some point ur just selecting for people w the time + resources to do it and (1/2)
page 1: every single thought, word vomit, brain dump, keep going til there’s nothing left in your head
page 2: organize
???
profit
not only do my cats get infinity food, water, naps, pets, but they also get to live in peace never having to learn a thing about AI 🥲
if i could time travel i know im supposed to unalive a dictator …..but also it would be so cool to see the inside of the mughal palace at the height of its glory
Niba 🎬 retweeted
Anthropic with every industry:
We're introducing Claude for Teachers: free access to premium Claude capabilities for verified K-12 educators in the US, with a library of teaching skills and a direct connection to evidence-based curricula, mapped to academic standards in all 50 states.
claude.com/solutions/teacher…
Niba 🎬 retweeted
Buddhists have been studying the mind for 2600 years. I'd like to help get them more involved in AI safety.
Today I'm announcing a new project to do just that.
Learn more at engagedbuddhists.ai, or in our announcement post below ⬇️
