What shipped & dropped across AI labs, today.
Today
Wed · Jul 22 · 6 postsAsk Claude about the Anthropic Economic Index
Economic A research agenda for the Economic Futures Research Fund
Cursor Router
Auto mode is now powered by Cursor Router. Cursor Router is our intelligent model router. It analyzes each request and sends it to the right model for the job. Frontier models handle work that demands them. Price-efficient models handle the rest. #Optimization modes Select Auto, then choose how the router optimizes: Intelligence: Frontier quality, matching the most expensive and powerful models that might be out of reach for daily use. Balance: Strong quality, matching the frontier models that most people like to daily drive. Cost: Good quality, reaching the highest available intelligence while optimizing token spend. Balance and Intelligence bill at the routed model’s rate. Each mode moves you along the cost-intelligence pareto frontier. #Admin controls Admins can enable the router per team or group, restrict which optimization modes members can use, set the default mode, and allow or block underlying models. Cursor Router is available across desktop, web, iOS, CLI, and our SDK. It is on by default for Teams plans. Enterprise admins can enable it from the dashboard. Learn more in our announcement and docs. Improvements↓↑ Per-request classification by task type and complexity Optimization modes: Cost, Balance, and Intelligence Admin controls: per-team and per-group enablement, mode restrictions, default mode, and model allow and block lists Routed model can be displayed or hidden (hidden by default) Soft and hard enforcement options for standardizing on Auto Grok 4.5 required as a price-efficient routing option
New Copilot usage metrics impact dashboard
We’ve released a new Copilot metrics impact dashboard for enterprise administrators and organization owners. This dashboard helps you tell a deeper Copilot impact story by sharing not just who is…
How Outtake built a cyber investigator on Claude
How Outtake ensures multi-hour agent sessions stay on track to uncover attack network operations
Building verification loops in Claude Code with skills
How to turn your manual checks into skills, so Claude closes its own feedback loop.
Yesterday
Tue · Jul 21 · 5 postsAnthropic is donating another $20 million to Public First Action
How Anthropic secures its AI-native software development lifecycle
Anthropic Deputy CISO, Jason Clinton, details how the Security Engineering team secures a SDLC that has AI authoring 80% of merged code.
Grok for Outlook
Use the Grok add-in for Microsoft Outlook to summarize long email threads, draft replies in your voice, and keep your inbox organized.
Gemini 3.6 Flash is now available in GitHub Copilot
Gemini 3.6 Flash, Google’s latest Flash model, is now rolling out in GitHub Copilot. It is designed for web and app development, coding, and longer-horizon agentic tasks. It has configurable…
How Datadog built a “universal machine tool” for Claude Code
Datadog has an agent write specifications for a deterministic kernel to write application code
Monday
Mon · Jul 20 · 6 postsAI credit pools for cost centers in the billing UI
You can now manage a cost center’s AI credit pool directly in the billing UI where you create and edit cost centers. Previously, you could only manage this through the…
Apply for Anthropic’s AI for Science rare disease research grants
Copilot users can now see AI credits used per billing cycle
Copilot Business and Copilot Enterprise users can now see how many AI credits they’ve used this billing cycle, even without an individual budget. Find this on your GitHub Copilot usage…
Run Ray on TPU, Part 1: The foundations
Ray 2.55 introduces official, first-class support for Google Cloud TPUs, enabling developers to run distributed Python workloads on Google's accelerators using the familiar Ray task-and-actor APIs. To handle the strict networking requirement of keeping multi-host TPU "slices" together over their Inter-Chip Interconnect (ICI), the KubeRay Operator on GKE automatically provisions and labels the underlying hardware layout. Ray Core utilizes these labels via its slice_placement_group() primitive to atomically reserve complete slices, allowing developers to deploy jobs through KubeRay, Ray Train, or Ray Serve simply by declaring a hardware topology (like "4x4") without writing custom placement code.
Working at the frontier: How Rakuten builds agents overnight with Claude Fable 5
Why Rakuten thinks Claude Fable 5 is a step change in model intelligence, transforming how long-running work is done.
Grok for Excel
Use the Grok add-in for Microsoft Excel to ask questions in plain English, write formulas, and run scenarios without leaving the workbook.
Friday
Fri · Jul 17 · 3 postsGitHub Copilot app now available in the usage metrics API
The Copilot usage metrics API now reports the GitHub Copilot app usage in the enterprise and organization 1-day and 28-day reports. This gives enterprise and organization admins visibility into the…
Repository-level GitHub Copilot usage metrics generally available
The Copilot usage metrics REST API now reports repository-level activity. Two new endpoints return a daily, per-repository breakdown of pull request activity for Copilot coding agent and Copilot code review.…
Copilot code review: Customization and configurability improvements
Copilot code review now utilizes a firewall, custom setup steps, and independent runner configurations. It now reads custom instructions from the head branch to allow for easy testing and validation…