Anthropic has released Claude Tag in research preview for Slack, an AI agent that maintains persistent memory across channels and can proactively participate in team conversations. Available to Claude Enterprise and Team customers, it differs from existing Slack integrations by learning organizational context over time and sharing a single identity across team members.
GitHub has released the redesigned terminal interface for GitHub Copilot CLI to general availability. The update, previewed at Microsoft Build 2026, introduces a tabbed layout for working with GitHub directly from the command line.
Anthropic's Claude services went offline on June 23 at 10:19 AM ET, affecting most models including Sonnet and Opus across all platforms except Claude for Government. The company deployed a fix by 10:53 AM ET, ending an outage that lasted approximately 35 minutes.
Mistral AI released OCR 4, a compact document extraction model that returns bounding boxes, block classification, and inline confidence scores alongside text. The model supports 170 languages, scores 85.20 on OlmOCRBench, and is priced at $4 per 1,000 pages via API ($2 with batch discount) or $5 per 1,000 pages through Document AI.
Google's Workspace Studio Flows is now available to Google AI Pro ($20/month) and Ultra ($100/month) subscribers, bringing AI-powered email filtering to Gmail. The service processes up to 2,000 emails monthly on Pro tier and 10,000 on Ultra, potentially limiting utility for high-volume users receiving thousands of messages weekly.
Google has expanded the Gemini overlay plus menu on Android to include six tools: Videos, Music, Canvas, and Guided Learning join the existing Images and Personal Intelligence options. The update, rolling out in Google app version 17.32, allows users to access most Gemini features from anywhere on Android without opening the full app.
Trail of Bits launched Patch the Planet, a security initiative using OpenAI's GPT-5.5-Cyber model to find and fix bugs in critical open-source projects. The first week produced 64 pull requests and 51 issues across 19 projects including cURL, Python, PyPI, and Sigstore, with 37 patches already merged.
Sakana AI released Fugu, an orchestration language model that routes tasks across multiple AI providers to reduce vendor lock-in risks. The Japanese AI firm positions Fugu as a solution to enterprise dependency on single monolithic AI APIs.
Baidu has released Unlimited-OCR, a 3 billion parameter model for optical character recognition and document parsing. The model supports single-page and multi-page document processing with a 32,768 token context window and runs on NVIDIA GPUs using bfloat16 precision.
Z.ai released GLM-5.2 on June 16, 2026, the first open-weight model to match proprietary models like Claude Opus 4.8 on agent benchmarks. The MIT-licensed model closes the performance gap to 6.8 months behind frontier labs, down from expected 9+ months as compute scales.
Tencent is testing an AI assistant called Xiaowei in Weixin, the Chinese version of WeChat, which has over 1.4 billion monthly active users combined with WeChat. Users can interact with Xiaowei through text or voice, communicate with friends, and launch mini-programs within the app.
Poolside has released Laguna M.1, a 225B total parameter Mixture-of-Experts model with 23B activated parameters per token, designed for agentic coding tasks. The model scores 74.6% on SWE-bench Verified and 63.1% on SWE-bench Multilingual, released under Apache 2.0 license.
The White House ordered Anthropic to restrict exports of its Mythos and Fable AI models last Friday, citing national security concerns. Anthropic pulled both models offline within 90 minutes of the Commerce Department directive, marking the first major test of AI export controls.
The US government forced Anthropic to withdraw its Fable 5 and Mythos 5 models, citing national security concerns after Amazon researchers allegedly discovered a method to bypass Fable 5's safety guardrails. Cybersecurity researchers have signed an open letter opposing the ban, with Anthropic noting similar vulnerabilities exist in competing models.
GitHub has released technical details on Qubot, an internal analytics agent powered by GitHub Copilot that enables employees to query company data using natural language. The agent represents GitHub's implementation of AI-assisted data analysis for internal operations.
GitHub has built Qubot, an internal analytics agent powered by GitHub Copilot that allows employees to query company data using natural language. The project represents GitHub's approach to building domain-specific AI agents for data analysis tasks.
Amazon Web Services launched Web Search on Amazon Bedrock AgentCore, a fully managed web search capability that gives AI agents access to tens of billions of documents without requiring external search APIs. The service, now generally available, runs entirely within AWS infrastructure and refreshes its index within minutes of new content appearing online.
Joanna Stern, former Wall Street Journal tech columnist, tested Apple's rebuilt Siri in iOS 27 for one week and reports substantial improvements. The assistant now pulls context from Messages, Calendar, and voicemail to deliver personalized responses—though limitations remain in current beta.
Google has updated Gemini Live to access past conversation history through Memory and Connected Apps. The feature, currently available in English in the US, allows the voice assistant to reference previous chats and information from YouTube, Workspace, Utilities, and image generation tools during conversations.
Amazon Web Services made its AgentCore harness generally available, reducing production AI agent deployment to two API calls: CreateHarness and InvokeHarness. The managed service handles sandboxed execution, memory, tool integration, and observability, eliminating infrastructure setup for teams building LLM agents.