Researcher at @mcgillu combining AI and neuroscience. Also on Bluesky (@tyrellturing.bsky.social) and Mastodon: @tyrell_turing@fediscience.org.
Montréal, Québec
Joined April 2013
- Tweets23.1K
- Following1.8K
- Followers16.1K
- Likes54K
Pinned Tweet
1/15) What could drive AI agents to cooperate with each other, even if there is no chance for reciprocity or pay back? 🤔
🧵 Our team at Google, Paradigms of Intelligence, uncovered new paths to cooperation and a new game theory for foundation models 👇
Blake Richards retweeted
Couldn't be happier to announce my new paper on defining compositional structure — and generating it! I see this project as the culmination of my PhD, as it unifies all my interests: compositionality, compression, structure, and open-endedness. More below! arxiv.org/abs/2610.20922
Blake Richards retweeted
1/ We discovered something totally unexpected about the neural structure of consciousness that overturns decades of thinking on its biological basis and brings us significantly closer to the recipe for generating conscious awareness.
New preprint: biorxiv.org/content/10.64898… (led by @HesseJanis, with @FrankLanfranchi, @_conor_mcgrory, @Yuelin_Shi, and @EngelTatiana)
Blake Richards retweeted
The olfactory cortex doesn't store smells. It reshapes the space they live in. We wrote a fun review on representational manifolds in PCx with @RainerFriedri12 → doi.org/10.1016/j.conb.2026.…
Blake Richards retweeted
1/5
Continuous diffusion may be better at discrete reasoning than standard sampling suggests.
We took SRM's MNIST-Sudoku Hard checkpoint (arXiv 2502.21075) and changed ONLY the sampler:
❗ 0.8% → 65.7% valid boards solved, without retraining. Why does standard sampling fail?🧵
Blake Richards retweeted
Proud to share new work from our Paradigms of Intelligence team toward AI designed for decentralized hardware, where low-power communication and fault tolerance are critical requirements. Using Neural Cellular Automata (NCA), @mayalen_etc and her co-authors show that spatial locality is not a barrier to complex multi-step reasoning.
Read the full paper on arXiv: arxiv.org/abs/2609.36126v1
Blake Richards retweeted
Our recent work presents a 1-gram wireless mesoscope for cortex-wide imaging at single cell resolution in naturally behaving mice, published in @Nature . It's interesting to see large-scale neural activities in multiple mice within a very natural habitat. rdcu.be/iGFq4KbdWuza
Blake Richards retweeted
Paper out!
arxiv.org/abs/2610.19487
"How long is the coast of a neuron?
The point neuron — ubiquitous in theoretical neuroscience, hardwired into every neural net — is a ruler. Coarsen it and dendritic computation looks additive. Refine it and a readout gains capacity for free."
Preprint soon. Here is the trailer!
youtube.com/shorts/mkQJfYQDY…
Check out our new model for creating realistic motion in a virtual rat during RL!
I think of this as providing the "brainstem and spinal cord" for NeuroAI researchers who want to simulate realistic neuroscience experiments.
Congrats to @aidan_sirbu for getting this out!
1/14) Virtual animals are a long-standing goal in neuro-AI, but they should actually behave like real animals.
We present SCAMPER: a virtual rat for RL studies that moves naturally during both exploration, and goal-directed behavior.
Project page: aidansirbu.com/SCAMPER-site/
Blake Richards retweeted
Friends in San Francisco: Join me at @longnow on October 22 for a conversation on interdependence with @baratunde.
We'll explore cooperation as the core governing principle of nature and synthetic minds, why intelligence is inherently collective, and what this means for our shared future.
Get tickets here: eventbrite.com/e/on-interdep…
How do we co-create a future with tech that actually serves humanity and the planet? On Oct 22, join @baratunde (creator of How to Citizen & Life With Machines) alongside Google VP @blaiseaguera for "On Interdependence": a 10,000-year look at civic action, Indigenous wisdom, and human survival.
More information and tickets here: na2.hubs.ly/H082rvS0
Blake Richards retweeted
DeltaNet-PC is out! It turns out that predictive coding offered a native mechanism for test-time memorization of sequential data all along. The crucial insight was to treat updates to the fast associative memory as an inferred latent variable, not a long term learned weight.
Blake Richards retweeted
Thank you Rocket Drew @rocketalignment for hosting me on The Information about world models and interp!
Rocket is a precise & careful technical interviewer who is a pleasure to talk to, and who let me go into a technical rabbithole while keeping the convo accessible!
🚀 AI Deep Dive Episode 6: Are We Building World Models The Wrong Way?
@rocketalignment and World Mechanics co-founder @soniajoseph_ explore how models represent physics, whether interpretability could improve training and why language alone may not be enough for physical AI.
00:00 Intro
01:39 What counts as a world model?
07:12 Looking inside models for physical understanding
11:28 Interpretable training and the bitter lesson
19:24 Physics, language and understandable explanations
28:07 Surgical robots and manufacturing
31:10 Language models and recursive self-improvement
38:13 JEPA and specialized world models
45:25 What video models actually learn about physics
Blake Richards retweeted
Mila's annual supervision request process is now open to receive MSc and PhD applications for Fall 2027 admission! For more information, visit mila.quebec/en/prospective-s…
Blake Richards retweeted
Over 400 researchers and I have signed this open letter calling for the AI and security communities to come together to improve alignment and containment approaches in order to protect our critical infrastructure and digital systems.
ai-security-letter.org/
Blake Richards retweeted
Over 400 academics in 30+ countries signed our open letter calling for immediate changes in AI development and deployment practices to ensure that cybersecurity incidents do not continue to occur, as future incidents will be far more damaging to our critical infrastructure.
I signed! This is some common sense stuff.
Blake Richards retweeted
We've released our first blog post on interpretability as a critical ingredient for scaling physical intelligence in a way that is data efficient, generalizable, and verifiable!
Blake Richards retweeted
1/ Why does predicting in latent space (JEPA, CPC, SimCLR...) work so well on messy data, with changing lighting, camera angles, and busy backgrounds? The usual answer: it can ignore nuisance. But that answer holds a conundrum.
Blake Richards retweeted
1/9 New paper led by Michael Hauri & Peter Buttaroni: “Learning Commute-Time-Preserving World Models for Planning.” We show how to shape a world model's latent space so its distances mean something for planning.🧵
📄 arxiv.org/abs/2610.01373
🌐 ctwm-website.github.io/
PSA for all the folks submitting AI generated papers to ML conferences:
ACs are now being given the power to desk reject papers for being incomprehensible, and we will use this power.
Make sure your paper is actually human parsable, FFS!
Blake Richards retweeted
the hardware lottery is also the hardware curse.
with our team, we've been pushing the boundaries of what fully local neural cellular automata models achieve on visual reasoning.
we believe this opens the door to a swathe of efficient, robust, and scalable hardware designs.