Anthropic opens Singapore office with AI safety MOU
The opening adds a local presence as Singapore seeks closer AI cooperation.
Anthropic
Anthropic news covering Claude, coding tools, AI research, safety, company strategy, and policy, with clear context and links to the evidence.
Fable 5.5 leak faces correction after LiteLLM entry fuels launch rumors
A disputed LiteLLM listing sparked Fable 5.5 launch rumors. Here is what the original change and official model documentation actually establish.
Anthropic replaces Git with Scion as daily merges surge
The engineering account raises a practical question for teams adopting coding agents. Which parts of the delivery pipeline should change first?
Workday connects Claude and Sana for business workflows
The partnership links Claude with Workday agents, while rollout stages determine what customers can use.
Claude brings direct editing to Google Docs Sheets and Slides
Claude adds direct editing inside Google files and through separate connectors. The routes differ in setup, access, charts and enterprise limits.
Anthropic reports stronger distillation defenses
Anthropic reports reduced reasoning extraction alongside developer tradeoffs.
Anthropic tests AI replication on everyday connected hardware
Anthropic examines AI replication under simplified conditions and identifies limits in small models.
Claude Code 2.1.296 adds finer controls for delegated work
New controls target worker context, workflow models and large file reads. Configuration details still need documentation.
Claude Managed Agents adds dynamic workflows for parallel jobs
The new beta coordinates work in the background. Budgets, permissions and coverage checks matter before expanding a workload.
Claude Startups reaches capacity on Team and API offers
Anthropic says it will review all startup applications again while claimed offers remain usable. Founders should check their account before relying on a grant.
Claude Dashboards and Motion move into everyday reporting
Claude Dashboards and Motion bring new ways to build reports and explainers, with practical limits around access, sharing and rollout.
Anthropic launches free code scans and infrastructure defense program
A closer look at OSS Scanner eligibility, enrollment, accuracy evidence and the review work that remains with maintainers.
Anthropic Is Banning Abuse of Claude
What the Claude abuse rule covers, how conversation ending differs from account restrictions, and where to appeal.
Anthropic commits $150 million to expand Claude access for federal science
Claude credits and technical support will help federal agencies pursue Genesis Mission research.
Claude Haiku 5.5 has launched and the early demos show its limits
Claude Haiku 5.5 has launched. Our video examines early creations, visible failures and the costs that matter when choosing the smaller model.
Claude Opus 5.5 motion graphics show why the setup matters
Creator examples span quick product explainers, long rendering jobs and editable animal models. The brief and tools shape the result.
Anthropic reports 6,157 security findings sent to maintainers
Anthropic’s October 2 snapshot reports more findings sent to maintainers, with separate review and patch counts that need careful interpretation.
Anthropic commits $100 million to Claude Frontier Academy
Claude Frontier Academy aims to train 10,000 enterprise engineers through practical assessments and a workplace residency.
Claude Code update repairs recovery and tool execution controls
Claude Code 2.1.288 addresses interrupted sessions, repeated MCP calls and restored model restrictions. Teams still need to test their configured controls.
BootLoops brings reusable scientific tools to AI research
Matthew Schwartz releases a toolkit built around checkable calculations and human direction
Anthropic study finds robot costs limit physical automation
New research compares demonstrated robot capability with deployment economics
Claude for Government reaches general availability
Anthropic moves its government product beyond beta with separate early access additions
Claude Code mods open deeper customization through TypeScript
Anthropic launches mods for changing coding workflows and interfaces
Claude Code adds provider controls and clearer model fallback
September releases improve provider restrictions model fallback and gateway usage accounting