Strengthening safeguards with US CAISI and UK AISI
Government red-teamers tested Constitutional Classifiers on Claude Opus 4 and 4.1, uncovering vulnerabilities that strengthened our safeguards.
Daily digest / 19 October 2026
AI news digest — 19 October 2026
Expanding Project Glasswing
We’re extending Project Glasswing to approximately 150 new organizations in more than fifteen countries
Partnering with Accenture on embedded evaluation
We’re partnering with Accenture on independent evaluation of frontier AI—part of our recent commitment to embed evaluators at Anthropic. Both we and Accenture expect to invest at least $1 billion to build capacity in this area over the next five years.
Responsible Scaling Policy Version 3.0
An update to Anthropic's policy to mitigate catastrophic risks from AI
Cooking with Codex: Ep. 1
In this episode of Cooking with Codex, an OpenAI dot helps plan a ramen recipe and organize prep in ChatGPT Space, with GPT-Live guiding the cooking process. Also covered: Codex cloud environments, which keep tasks running after a laptop is closed, and new Codex CLI features i…
Muse Spark Open leads U.S. open models: Reflection releases weights for Beam, a much-anticipated model
Grok Voice 3.0 leads voice-to-voice benchmarks. Nvidia 3.5 Super VL is faster, more capable than Lightning. DeepSeek-V-4.1-Flash pushes strong update. Meta amends controversial Muse scraping terms.
Cutting our inference bill by 68% by being obsessive about optimization
When our AI code reviewer's cost per run doubled, we dug into the token usage to find out why. Learn how we used prompt caching and intelligent model routing to slash our inference bill by 68% while speeding up reviews.
How to continue a GitHub Copilot app session in VS Code
Learn how to move an agent session from the GitHub Copilot app into VS Code without losing the conversation, code changes, or working context. The post How to continue a GitHub Copilot app session in VS Code appeared first on The GitHub Blog .
Why teaching AI about symmetry could make prenatal diagnosis more reliable
A model developed by researchers at MIT and Boston Children's Hospital adjusts scans on the fly to track fetal movement, potentially sparing families repeat hospital visits.
Introducing Clairic, a new System One class of extraction models
Clairic is a new model from Cloudflare that is a System One extraction modeI, capable of extracting, computing, estimating, and forecasting data from unstructured inputs.
Accelerating Mamba-2 with High-Performance Forward Pass Kernels in NVIDIA CUDA Tile
NVIDIA CUDA Tile (cuTile) is a Python-based DSL and parallel programming model for NVIDIA GPUs that automatically leverages advanced hardware capabilities to... NVIDIA CUDA Tile (cuTile) is a Python-based DSL and parallel programming model for NVIDIA GPUs that automatically le…
GenRecs Tokenization: Transforming Ads Recommendation at Meta with Stable Model Vocabulary
Generative Recommenders (GenRecs) represent a paradigm shift from scoring every ad to reasoning about ad content and user preferences together in order to predict the best outcome for advertisers. Leveraging the world knowledge built into LLMs makes ad matching more intelligen…
The Thin Orchestrator: How We Made Our AI Agents More Reliable by Giving the Router Less to Do
Sometimes, giving an AI agent all of the knowledge it needs is the wrong approach. At Meta, we’ve seen notable improvements in efficiency and accuracy by keeping the orchestration layer of our data pipeline agents deliberately thin: It routes work and enforces reliability chec…
Thomson Reuters on Building Trust in Legal AI | OpenAI
Bringing legal expertise and frontier AI together starts with trust. For Thomson Reuters’ Camillo Speroni, that means output lawyers can verify, cite and put their name on. Camillo discusses Thomson Reuters’ work with OpenAI to bring HighQ and Thomson Reuters CoCounsel Legal i…
Thomson Reuters on Building Trust in Legal AI | OpenAI
Bringing legal expertise and frontier AI together starts with trust. For Thomson Reuters’ Camillo Speroni, that means output lawyers can verify, cite and put their name on. Camillo discusses Thomson Reuters’ work with OpenAI to bring HighQ and Thomson Reuters CoCounsel Legal i…
Inside Prometheus: Meta’s Journey to a Gigawatt-Scale Supercluster
With the recent release of the Muse family of foundational models, we want to take you inside the infrastructure that has made it all possible. The first phase of Prometheus, our gigawatt-scale supercluster, is already complete and we’re working toward scaling it to comprise ~…
Import AI 477: In which I tell the IMF the Singularity has begun
What do you think?
Advancing coding in low‑resource programming languages
Learn how OpenAI is improving ChatGPT across low-resource programming languages, with new evaluations for software engineering, SQLite, and COBOL migration.
New course teaches how AI models run on modern accelerators
'Dataflow Architectures for AI' is a 6-course specialization from IBM Research that compares leading accelerator architectures through hands-on labs.
Speech to text for content creators: A workflow guide for captions and repurposing
See how speech to text for content creators improves their workflows and accelerates production across captions, show notes, clips, LLM analysis, and more
Announcing the Chroma Awards Season 2
Today we’re opening the Chroma Awards Season 2, a global competition to find the world’s catchiest ad.
TorchNeuron: running HF models on AWS Trainium is easier than ever
Together Post-Training: frontier-grade model shaping, the way that fits you
Post-train open LLMs like Qwen, GLM and Kimi into custom models you own, with SFT or RL. You write the training loop. Together runs the hard infrastructure.