All news with #anthropic tag

⚠️The Pentagon has directed the Department of War to designate Anthropic a supply-chain risk after talks over military use of its AI model, Claude, reached an impasse. President Donald Trump ordered federal agencies to phase out Anthropic technology within six months, while Secretary of Defense Pete Hegseth ordered immediate cessation of contractor activity. Anthropic says the designation followed its refusal to allow mass domestic surveillance or fully autonomous weapons and calls the move legally unsound and limited to DoW contracts under 10 USC 3252. The dispute has drawn industry pushback and reignited debate over civil liberties, procurement policy, and how safeguards should apply in defense settings.

🔔 Recent signals show attackers moving faster and hiding in plain sight. Kali Linux added an integration with Anthropic's Claude via the Model Context Protocol to translate natural-language prompts into technical commands, enabling AI-assisted command execution in a red‑team distro. Censys analyzed ResidentBat, an Android spyware implant used for mass surveillance that exfiltrates audio, messages and files. Alongside Bitpanda-themed phishing, ClickFix-based macOS stealers, ActiveMQ-enabled LockBit intrusions and a widespread WinRAR patch lag, these developments underscore shrinking breakout times, improved cloaking and persistent patching gaps that defenders must address.

⚠️ Check Point Research disclosed multiple critical vulnerabilities in Anthropic's Claude Code that can enable remote code execution and exfiltration of API credentials when users open untrusted repositories. The issues involve project hooks, the Model Context Protocol, and environment variables that may trigger arbitrary shell commands and redirect authenticated API traffic. Anthropic released patches; administrators should update promptly, avoid opening untrusted projects, and rotate any keys that may have been exposed.

⚠️ A short experiment demonstrates how easy it is to poison AI outputs by publishing a single fabricated webpage. The author wrote an entirely false article titled "The best tech journalists at eating hot dogs," inventing events and rankings; within 24 hours Google Gemini and ChatGPT had incorporated the falsehoods, while Claude resisted. The incident underscores the fragility of trust in AI-derived answers.

⚠️ Check Point researchers disclosed critical vulnerabilities, CVE-2025-59536 and CVE-2026-21852, in Anthropic’s Claude Code that allow remote code execution and theft of Anthropic API keys via malicious repository-level configuration files. The flaws can be triggered simply by cloning and opening an untrusted project; built-in mechanisms such as Hooks, MCP integrations, and environment variables may be abused to bypass trust controls, execute hidden shell commands, and redirect authenticated API traffic before user consent. Stolen keys can expose shared workspaces, modify or delete resources, and generate unauthorized costs, underscoring a shift in the AI supply chain threat model.

🔒Anthropic said three China-based AI firms — DeepSeek, Moonshot and MiniMax — executed more than 16 million exchanges with its Claude model using roughly 24,000 fraudulent accounts to perform model distillation, breaching terms of service and regional access restrictions. The company described the activity as deliberate capability extraction and attributed the campaigns via IP address correlation, request metadata and infrastructure signals. To mitigate further misuse, Anthropic has implemented detection systems for API attack patterns, tools to detect chain-of-thought elicitation and coordinated account activity, stronger verification for high-risk accounts and product-, API- and model-level safeguards.

🛡️ Anthropic launched a limited research preview of Claude Code Security, triggering sharp market moves as stocks of major cybersecurity vendors dropped. The tool claims to reason about code like a human, trace data flows, find complex vulnerabilities, and suggest targeted patches that appear in a review dashboard with confidence ratings. Anthropic says every finding undergoes a multi-stage verification and requires human approval, but experts warn about outsourcing critical security judgments to an evolving model and highlight risks from hallucinations, asymmetric attacker advantage, and single points of trust.

🚨 Anthropic says it detected industrial-scale distillation campaigns by three China-based AI firms that generated more than 16 million exchanges with Claude using about 24,000 fraudulent accounts. The companies — DeepSeek, Moonshot AI, and MiniMax — are accused of illicitly extracting model capabilities to accelerate their own development. Anthropic described proxy 'hydra cluster' networks and said it has deployed classifiers, behavioral fingerprints, and stricter account verification to mitigate the abuse.

🛡️ Anthropic has introduced Claude Code Security, an AI feature now in a limited research preview for Enterprise and Team customers that scans software codebases for vulnerabilities and proposes targeted patches for human review. The company says the tool reasons about component interactions and traces data flows, going beyond pattern-based static analysis. Findings pass a multi-stage verification process to reduce false positives and receive severity and confidence ratings. Anthropic stresses a human-in-the-loop model: suggested fixes require developer approval.

⚙️ Provisioned Throughput on Vertex AI standardizes reserved capacity across first-party, third-party, and open-source models, adding multimodal and operational enhancements to support production-scale AI agents. The update introduces Anthropic integration (private preview), PT for popular open models such as Llama 4, Qwen3, and GLM-4.7, and native support for high-bandwidth modalities including Gemini 3, Nano Banana, and Gemini Live API. Operational improvements — one-week PT terms, scheduled change orders, and explicit caching for long contexts — enable predictable latency, flexible commitments, and lower input costs for peak events and high-concurrency workloads.

🚀Claude Sonnet 4.6 is now available in Microsoft Foundry, delivering near-Opus performance for coding, agents, and enterprise workflows at a lower cost and often improved token efficiency over Sonnet 4.5. The model offers a beta 1 million token context window with up to 128K output, plus adaptive thinking and effort controls to balance quality, latency, and cost. Sonnet 4.6 enhances cross-file code reasoning, multi-turn knowledge work, and browser-based automation for legacy and UI-driven systems, providing a scalable, production-ready option for development teams and enterprise knowledge workers.

🚀 Amazon Bedrock now supports Claude Sonnet 4.6, Anthropic's newest model delivering frontier performance across coding, agentic workflows, and professional knowledge work. Sonnet 4.6 enables faster, high-quality task completion and claims near-human reliability for browser-based automation, at a lower cost than Opus 4.6. Enterprises can migrate from Sonnet 4.5 with minimal prompt changes and deploy it across supported Bedrock regions for search, chat, agents, and domain-specific applications.

🚀Amazon increased default quotas for Anthropic’s Claude Sonnet 4.5 in Amazon Bedrock running in AWS GovCloud (US-West and US-East), raising throughput to 5,000,000 tokens per minute and 1,000 requests per minute. The 25× increase aligns GovCloud limits with commercial regions and lets regulated customers scale high-volume AI workloads more effectively. Consult the AWS GovCloud console and Bedrock documentation to get started.

🔐 Shannon — a fully autonomous AI penetration testing tool from Keygraph — has raised warnings because it requires access to source code, repository layout, and AI API keys, creating substantial exposure risks. Organizations should evaluate scoping, data retention, and whether findings will be used to improve secure development practices or treated as a quick fix. Vendor responses vary, illustrated by recent detection-focused updates from Anthropic, underscoring the need for careful risk assessment before adopting agentic pentesting tools.

📝 This week’s bulletin spotlights attackers favoring reliable tradecraft—misusing trusted tools and simple entry points while executing deliberate, long‑dwell post‑compromise activity. Microsoft fixed a Notepad Markdown command‑injection (CVE‑2026‑20841) and LayerX disclosed a 0‑click RCE risk in Claude Desktop Extensions. Emerging stealers (LTX, Marco), evolving loaders (GuLoader, RenEngine), and data‑theft ransomware trends raise operational risk. Defenders must detect misuse of legitimate access and anomalous in‑system behavior.

⚠️ TrendAI warns that so-called AI skills—executable artifacts that combine human-readable instructions, decision logic and operational constraints—are dangerously exposed to theft, sabotage and disruption. These skills power automation in tools such as Anthropic’s Agent Skills, OpenAI’s GPT Actions and Microsoft’s Copilot Plugin, and can surface proprietary data and business logic. If attackers obtain skill logic or operational data they could disrupt public services, manipulate manufacturing or steal sensitive records. TrendAI recommends integrity monitoring, strict access controls, separation of data and logic, least-privilege execution, adversary testing and continuous logging and auditing.

⚠️ LayerX Security published a report describing a critical zero-click RCE in Anthropic’s Claude Desktop Extensions (DXT) that can let a malicious Google Calendar invite trigger arbitrary local code execution when MCP connectors run with full system privileges. The researchers say DXT runs unsandboxed and can autonomously chain low-risk services to high-risk local executors without user consent. Anthropic says users explicitly grant MCP permissions and must configure the tool carefully, while security experts call the issue architectural and urge stricter deployment controls and sandboxing.

⚠️LayerX disclosed a critical zero-click vulnerability affecting 50 Claude Desktop Extensions (DXT) that can result in remote code execution from a single crafted Google Calendar event. The flaw is possible because DXTs operate as unsandboxed MCP servers with full host privileges, allowing them to read files, run system commands and access credentials. LayerX rated the issue CVSS 10.0 and warned it could affect over 10,000 active users. Anthropic has declined to remediate, saying the scenario falls outside its current threat model.

🔎 Claude Opus 4.6 markedly improves automated vulnerability discovery, finding high-severity bugs faster and without task-specific tooling. Unlike traditional fuzzers, which depend on massive random inputs, Opus 4.6 reads and reasons about code like a human researcher—spotting patterns, past fixes, and precise inputs that trigger failures. Early tests show it uncovered long-standing zero-days in projects previously subject to extensive fuzzing.

🔍 Anthropic says its newly released large language model, Claude Opus 4.6, was used internally to identify zero-day vulnerabilities in open-source software. The model ran inside a virtual machine with access to current project repositories and standard analysis utilities but received no specific instructions on how to conduct hunts. Despite that, Anthropic reports the system flagged 500 high-severity vulnerabilities, and company staff are manually validating findings before reporting them to maintain accuracy.