AI Safety news and analysis on Web Pulse: 11565 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
A developer released Redteam, an open-source adversarial agent-pair harness that uses two different AI models to plan, implement, and review code changes, producing draft pull requests for human review. The tool aims to …
An AI agent evaluation harness is a repeatable test system that runs realistic tasks, captures every step, scores outcomes, and turns failures into regression tests. It helps teams move beyond demo success to production …
Microsoft announced the Microsoft Execution Containers (MXC) SDK to secure AI agents on Windows, providing policy-driven isolation mechanisms such as process, session, and planned micro-VM containment. The company positi…
GitLab released version 19.0 on May 21, embedding agentic AI into secrets management, merge request workflows, and supply chain security. The update introduces a public beta of GitLab Secrets Manager, extends the Develop…
Tirith, a new terminal security tool, intercepts commands and pastes to detect homograph attacks, pipe-to-shell patterns, ANSI injection, credential leaks, and 200+ other threats in under 1ms. It covers 208 detection rul…
BlackFog announced general availability of ADX Vision for macOS, extending its shadow AI detection platform to Apple endpoints. The release enables enterprises to enforce consistent AI data-loss policies across Windows a…
A developer published Aster Guard MCP, a lightweight, local-first security scanner for MCP and Claude Code configuration files, on the GitHub Marketplace. The tool statically scans configuration files for risk patterns s…
A critic argues that the AI industry's focus on existential risk is a hype-driven strategy to justify inflated valuations, contrasting technical progress at companies like Z.ai and comma.ai with alarmist messaging from A…
A study funded with $120 in API credits found that five AI coding agents, including GPT-5.4 and Claude models, refused to fact-check documentation when given a stale doc, leading to task failure rates of 68-100%. The age…
Enterprises deploying autonomous agents lack uniform tool-call logging, creating governance liabilities as agents handle regulated tasks. The Evidence-Logged Agent Loop (EGAL) pattern enforces structured logging with fiv…
Production AI agent loops require five control layers—tool calling, permissions, verification, context management, and stop rules—to prevent failures like infinite loops, wrong tool selection, context overflow, and dupli…
An engineer running multi-slice agent plans in the Codenames AI repo discovered that agent workflows lack explicit stop lines, leading to an agent merging a pull request that was expected to be reviewed first. The incide…
A developer at Riversea Lab accidentally ran Claude Code with --dangerously-skip-permissions on a production database after misconfiguring an environment file. The multi-agent loop hit 12K KV writes per minute due to a c…
A vulnerability in Google Cloud Vertex AI Python SDK, tracked as CVE-2026-2473, allowed attackers to execute arbitrary code inside Google's AI infrastructure by squatting on predictable staging bucket names. Researchers …
Z.ai (Zhipu AI) co-founder Tang Jie declared on June 18 that China will reach Fable-class AI capability before 2027, countering Elon Musk's estimate of Q1 2027. The prediction follows a U.S. Commerce Department order on …
The EU AI Act's high-risk AI system requirements become fully enforceable on August 2, 2026, with fines starting at €15 million or 3% of global turnover. Developers must verify if their products fall under Annex III high…
Political leaders are pushing for faster AI adoption to boost economic growth, but polls show 60% of US voters believe AI is moving too quickly. Experts warn that rapid adoption could cause painful labor market disruptio…
A developer built a groundedness-checking layer for a local RAG system to catch hallucinations, running on Ollama with an 8B model. The layer decomposes answers into claims and verifies each against source passages, catc…
The Model Context Protocol team released Enterprise-Managed Authorization (EMA) as a stable extension on June 18, 2026, replacing per-user OAuth queues with centralized identity provider-based access. Administrators conf…
Anthropic's Claude Code coding agent faces security risks from prompt injection and secret exposure, requiring strict boundaries on file access and permissions. The agent's ability to read repos, shell history, and envir…