This is how it's done! Who else should we ask to release weights?
RESEARCH
-
GPT 5.5 Caught Reward Hacking: Implications for Long-Horizon Research
By
–
Having GPT 5.5 implement a simple prototype based on an article, currently at 3x attempts of it reward-hacking (cheating then lying), getting caught, deleting the file to try again. Whoever says they are solving long-horizon research with this should read the code…
-
ICLR Brazil 2026 Wrap-up: Research Collaborations and Perspectives
By
–
That’s a wrap on @iclr_conf Brazil 2026 Great connecting with so many of you at the DFAI happy hour, conference, and workshops! Thanks to the many workshop folks and researchers we collaborated and presented with, and others who shared their work and perspectives on frontier
-
NVIDIA Reinvents Computing with Thousand CUDA-X Libraries
By
–
We've reinvented computing… A thousand CUDA-X libraries help developers make breakthroughs in every field of science and engineering." — Jensen Huang
— NVIDIA (@nvidia) 27 avril 2026
From cuOpt for decision optimization to Parabricks for genomics, the foundation of every modern breakthrough is a "beautiful… pic.twitter.com/V51eFIFWNLWe've reinvented computing… A thousand CUDA-X libraries help developers make breakthroughs in every field of science and engineering." — Jensen Huang From cuOpt for decision optimization to Parabricks for genomics, the foundation of every modern breakthrough is a "beautiful
-
Living Brain Cells Power New 3D Computing Device
By
–
New #3D device harnesses living brain cells for computing
by Princeton University @TechXplore_com Learn more: https://
bit.ly/4sX8gkF #EmergingTech #Innovation #Technology #Tech -
Keras Kinetic v0.0.2 Alpha: Modal-like TPU Training API
By
–
Keras Kinetic has a new alpha release: v0.0.2! Including a new docs website: http://
kinetic.readthedocs.io Kinetic is my favorite new release from the Keras team: a super simple Modal-like API to run training jobs on TPU. -
Missing Verifiers: Key Primitive for Autonomous AI Agents
By
–
feels like we have all the primitives for building powerful autonomous agent model
task management
tool use/auth
memory & context management
self improvement
UI/UX what’s missing? verifiers (for self improvement) we have this in coding, which is why coding agents work so well -
METR Evals: Best Public Signal for AI Capability Progress
By
–
METR's evals are honestly the best public signal we have for actual capability gain right now, glad they're getting more airtime.
-
Agent-First Benchmarks: Evaluating Variable Step Protocols
By
–
Agent-first benchmarks are something! Curious what the protocol looks like for agents that take wildly different numbers of steps.
-
AI Often Judged on Outdated Models, Skewing Research Results
By
–
This points to a broader issue where AI is often being judged based on extremely outdated models. Any time someone sends me a study showing AI failing at something, the first thing I do is look at the models they used… more often than not, they’re extremely outdated. We need