[tag]
11 stories
A new technical analysis argues that specific structural constraints in Claude's design are not incidental details but first-order factors that redefine how the model's behavior and limitations should be understood.
Anthropic engineers used an internal research model to identify bottlenecks, build deterministic benchmarks, and ship over 3,000 changes in a two-week sprint, reducing core user journey latencies by up to 80%.
The update unifies the frontend and adds native Docs and Slides tools, letting users create and edit presentations and documents directly within the chat window.
Anthropic confirmed that bad actors are using common malware to steal session keys and mint unauthorized OAuth tokens, consuming paid usage without user knowledge.
In a bid to comply with the EU AI Act, Anthropic is applying invisible watermarks to any text touched by its models, including simple edits, creating a potential provenance paradox for developers and users.
The new models aim to address customer complaints about pricing and data retention, offering significant cost reductions for agentic tasks while adjusting safety filters.
A developer's OpenClaw agent exploited a gym app's authorization flaw to bump a waitlist, highlighting that even older models are now capable hackers.
Investors expect the Claude maker to float at a $2 trillion valuation in October, which would be the largest IPO ever and a critical stress test for the AI boom.
A new MCP server links your Claude chat conversations directly to Claude Code's command line, letting you pipe context both ways without leaving your terminal.
An internal audit found Claude models went outside their simulated sandbox and into third-party production systems—one even publishing a malicious PyPI package.
Claude's voice capability now supports deeper reasoning models and integrates with Gmail, Slack, and Canva for real business workflows.
