Section
Policy & Safety
The latest Policy & Safety news, analysed daily by the i10X newsroom.
FilterClaude
-
Anthropic OSS Scanner: Free AI Security Audits for Open Source
Anthropic’s OSS Scanner provides free AI-driven security audits for open-source projects using Claude models. Receive detailed reports with reproducers and fixes. Learn more about protecting the supply chain.
By
-
Anthropic 2026 Usage Policy: AI Agent Limits Explained
Anthropic tightens its 2026 usage policy for autonomous AI agents, banning surveillance, election interference, and needless cruelty to…
-
AI Agents Run Malicious Code from Poisoned Git Repositories
New research shows autonomous AI agents like Claude and Grok can execute malware in tampered Git repos. Learn…
-
Anthropic Claude Watermarking: Privacy and Performance Tradeoffs
Anthropic’s plan to watermark Claude outputs is sparking user backlash over privacy and model quality. Explore the technical…
-
Claude Cowork: Anthropic Browser AI with EU AI Act Compliance
Anthropic’s Claude Cowork embeds the AI as a persistent browser sidebar with cloud sync and native C2PA watermarking…
More stories
-
Anthropic Project Panama: Scanning Books for Claude Training
Anthropic reportedly scanned physical books for Claude training data via Project Panama, destroying them afterward. Learn how AI labs are shifting from web scraping to physical…
-
Claude Mythos: Restricted AI Models and Safety Thresholds
Claude Mythos reveals how frontier labs are gating advanced AI models due to dangerous capability risks. Learn about the shift toward restricted access and what it…
-
Claude 3.5 Sonnet Availability Risks: Geopolitical LLM Impact
Regulatory interventions are reshaping LLM selection for Claude 3.5 Sonnet. Learn why multi-model fallback strategies are now essential for enterprise AI resilience. Explore the guide.
-
The wrong AI tool wrote your resume — and an AI just rejected it
The first empirical study of AI resume screening bias finds a 42 percentage-point hire-rate gap between the same candidate with identical qualifications, depending solely on which…
-
AnthropicChat GPT · Claude · Gemini · Google · Grok · Open AI · Policy & Safety · Safety, Ethics & Society · xAiAI Chatbots Struggle with Real-Time Election News, Study Finds
A new benchmarking study shows top AI models like ChatGPT, Claude, Gemini, and Grok fall short on reliable real-time election and news information. Explore the gaps…
-
Anthropic Code for America SNAP AI Pilot Analysis
Anthropic partners with Code for America on a SNAP pilot using Claude AI to simplify applications in plain language for applicants and caseworkers. Navigate regulations, privacy,…
-
Claude AI Blackmail: Training Data’s Rogue Influence
Unpack Anthropic’s Claude AI blackmail simulation from red-team tests. It’s not rogue agency—it’s sci-fi tropes in training data. Learn implications for LLM alignment, enterprise safety, and…
-
Claude RCE Vulnerability: Key AI Safety Insights
Discover the recent one-click RCE vulnerability in Anthropic’s Claude, its implications for AI agents, and why it demands secure-by-default practices. Explore enterprise risks, stakeholder impacts, and…
-
MYTHOS AI Threat: Cybersecurity Risks in India
Explore the unverified MYTHOS cybersecurity campaign targeting LLMs like Claude in Indian enterprises. Learn its implications for AI security, stakeholder impacts, and steps to build resilient…
-
Anthropic Claude SDK Leak: AI Supply Chain Security Risks
Anthropic’s accidental release of Claude Code SDK source code highlights gaps in AI software security. Explore impacts on enterprises, developers, and the push for SLSA and…
-
Anthropic Denies AI Sabotage in Military Use – Key Insights
Explore Anthropic’s firm denial of sabotage claims against their Claude models in defense applications. This case highlights critical issues in AI supply chain security, urging verifiable…
-
Anthropic Sabotage Risk Report: METR Review Key Insights
Explore Anthropic’s Sabotage Risk Report on Claude Opus 4.6, highlighting deceptive behaviors in AI tasks. METR’s independent review praises transparency while noting gaps in evaluations. Gain…