(1) Why the push into classified work now, when things are so hot? (2) when did you ready your models for the much more complex tasks you say classified work demands from AI? (3) What are the alignment risks associated with current OpenAI models that will be deployed by DoW?
SAFETY
-
Formal Verification for Neural Networks in Lean
By
–
Verification is a cornerstone for reasoning. Recent progress in mathematical reasoning and theorem proving has relied on LLMs + formal verification in Lean. But so far, Lean has focused on pure math, and lacks support for verification involving neural nets themselves. Applications include: 1. Certified robustness in neural networks, crucial in safety critical applications such as neural control. 2. Physics informed neural networks, which have been used to prove singularity problems related to the Millennium prize problem in fluid dynamics. 3. Theory related to neural networks such as approximation bounds, effect of quantization etc.
→ View original post on X — @animaanandkumar, 2026-03-01 01:04 UTC
-
TorchLean: First Fully Verified Neural Network Framework in Lean
By
–
We’re excited to release TorchLean which is the first fully verified neural network framework in Lean. The Lean community has largely focused on pure mathematics. TorchLean expands this frontier toward verified neural network software and scientific computing. With the recent release of CSlib, we see this as another step toward a fully verified ML stack. We support features: 1. Executable IEEE-754 floating-point semantics (and extensible alternative FP models) verified tensor abstractions with precise shape/indexing semantics 2. Formally verified autograd system for differentiation of NN programs Proof-checked certification / verification algorithms like CROWN (robustness, bounds, etc.) 3. PyTorch-inspired modeling API with eager-style development + export/lowering to a shared IR for execution and verification Project page: leandojo.org/torchlean.html Paper: [2602.22631] TorchLean: Formalizing Neural Networks in Lean Work done @Robertljg, Jennifer Cruden, Xiangru Zhong, @huan_zhang12 and @AnimaAnandkumar. #MachineLearning #ScientificComputing #Lean
→ View original post on X — @animaanandkumar, 2026-03-01 00:43 UTC
-
Laws Lagging Behind AI Capabilities: A Regulatory Challenge
By
–
The problem with "all lawful use cases" is that most of those laws were written way before we had any idea what these AI tools were capable of. Things that no one could conceive of are now mundane reality. Laws have always been way behind the technology, and at this point they
-
AI Safety Guardrails vs Usage Policies in National Security
By
–
Other AI labs have reduced or removed their safety guardrails and relied primarily on usage policies as their primary safeguards in national security deployments. We think our approach better protects against unacceptable use. In our agreement, we protect our redlines through a
-
Anthropic Rejects Supply Chain Risk Designation from Department of War
By
–
We do not think Anthropic should be designated as a supply chain risk and we’ve made our position on this clear to the Department of War.
-
AI Company Reaches Defense Department Agreement for Classified Deployment
By
–
Yesterday we reached an agreement with the Department of War for deploying advanced AI systems in classified environments, which we requested they make available to all AI companies. We think our deployment has more guardrails than any previous agreement for classified AI
-
OpenAI Department of War Agreement Establishes AI Usage Redlines
By
–
Our agreement with the Department of War upholds our redlines: – No use of OpenAI technology for mass domestic surveillance. – No use of OpenAI technology to direct autonomous weapons systems. – No use of OpenAI technology for high-stakes automated decisions (e.g. systems such
-
Gary Marcus Discusses AI-Related Scam in Substack Post
By
–
https://
open.substack.com/pub/garymarcus
/p/the-whole-thing-was-scam?utm_campaign=post-expanded-share&utm_medium=web
…