Congratulations!
Posts by Julian Togelius
And that should be @eugenevinitsky.bsky.social and no other Eugene
My PhD student @smearle.bsky.social very successfully defended his thesis today, in front of a committee consisting of @eugenerobinson.bsky.social @risi.bsky.social Tim Rocktäschel and @phillipisola.bsky.social . Next post: link to a recording of his defense talk
Large language models still have no ability to play video games well. @togelius.bsky.social, the director of New York University’s Game Innovation Lab, talks about why. spectrum.ieee.org/ai-video-gam...
Tonight we celebrate the Hungarian election. May this be a warning to corrupt and autocratic leaders everywhere.
I have complete faith in my health secretary, Robert Fungus Kennedy
For the STEM folks: Imagine if the editors at all of the top journals in your fields were second and third year graduate students.
A large proportion of the reviewers of large AI conferences (NeurIPS, AAAI etc) are PhD students. The conferences get so many submissions that there wouldn’t be enough reviewers otherwise. The graduate students often do a more thorough job than the professors.
Clever, though
New paradigm alert! 🎮
AgenticPCG
We combine classic PCG (Procedural Content Generation) algorithms with large language models for generating game levels. LLMs on their own are not good at level generation, but when given the right tools from our PCG toolbox they're killing it!
My team has been cooking! Turns out you can make LLMs creating great game levels… but only if you give them a whole bunch of existing metrics, generators, and parts of generators as tools. There’s lots more to do here. Agentic PCG for the win!
Great to have Luiz Chaimowicz visiting the Game Innovation Lab last week!
I think this interview about our recent position paper (with Georgios Yannakakis and Sam Earle) came out really well. Also, this picture (courtesy of Luiz Chaimowicz) is exactly what my days look like in the NYU Game Innovation Lab.
spectrum.ieee.org/amp/ai-video...
If one wanted to buy a barrel of crude oil and have it delivered to an office in Manhattan, how much would it cost? Asking for a hypothetical friend.
"Lessons from decades of game-playing machines suggest artificial general intelligence remains far away"
engineering.nyu.edu/news/why-ai-...
Nice summary by @nyutandon.bsky.social of our recent paper with @smearle.bsky.social, Graham Todd and Georgios Yannakakis
Never read posts from unknown numbers okay?
Hello, this post is regarding the status of your business loan
Play HAMM-3R! 🚀
FREE + SHORT (5-10 mins.)
>> hammer.gua-le-ni.com <<
An experimental game about tools and technologies made with @amidos2006.bsky.social, @demigiant.com and other friends at the @indigitalgames.bsky.social! 🔨
I played it but I was all too nice and was just doing what I was told. I must play it again and not be nice this time.
Surely, you need a steering committee!
DeepMind's RL team is hiring a research scientist: if you're passionate about RL, come work with us!
And if you know people who might be interested, please share:
job-boards.greenhouse.io/deepmind/job...
"[THING] has roots in bad 19th/early 20th century science" is the weed of humanist science/tech criticism. you smoke a blunt and it feels good, but the "insight" fades when the high does. Yes, you can find people with Bad Ideas because the foundations of modern science were made in a bad time
Blog post: Computers and me
togelius.blogspot.com/2026/03/comp...
A few weeks ago, I wrote a blog post about how I don’t know any maths. So the other day, I set out to write one about how I’m no good at using computers, or maybe rather that I’ve lost interest in computers? Don’t read this if you’re looking for a text with clear destination and coherent argument.
This is art
My talk from the Transformer: AI and the Future of Games conference a few weeks ago. I was sick, but the talk is still worth watching… I think
youtu.be/w2XDtIILews
the senators recoil in horror as a wounded Caesar rises from the floor and a second health bar appears on the screen
This development - people imitating language models - is honestly the most surprising aspect of how AI changes society. Like, why? Are people so eager to let go of their individuality?
Weird and cool