AI Safety news and analysis on Web Pulse: 23393 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
The TRENDS 6th Annual Conference in Abu Dhabi (2026) urged UAE CISOs and cloud architects to adopt sustainable security measures, including Zero Trust, data sovereignty, and on-device AI, to counter hybrid threats in a p…
A developer warns that AI-generated database migrations are outpacing human review, creating a 'reviewed without reviewing' risk where engineers approve plausible-looking SQL that may contain semantic errors. The post ar…
AccuKnox launched AgentZ on August 27, a zero-trust governance platform for production AI agents, citing Gravitee's 2026 State of AI Agent Security report that 48% of production AI agents lack adequate security controls.…
Sony Music and Warner Chappell filed a federal lawsuit against Anthropic, alleging a 'brazen campaign' of copyright theft that included co-founder Benjamin Mann personally torrenting over five million pirated books to tr…
A developer argues that as AI-generated code becomes more common, the ability to verify its correctness is becoming more valuable than writing it. Citing a 2025 Stack Overflow survey showing 84% of developers use or plan…
Anthropic shipped Claude Code v2.1.248 on Thursday with a new `--restricted` permission mode that disables shell execution, web fetching, and project-level settings files, addressing the TrustFall vulnerability class dis…
A new study introduces PLCBench, the first real-PLC hardware-in-the-loop framework for evaluating whether autonomous large language model (LLM) agents can convert network access to programmable logic controllers (PLCs) i…
A developer built Airlock, an open-source tool that audits MCP servers by comparing their declared tool annotations, such as readOnlyHint, with observed behavior. In tests, Airlock detected all five planted dishonest beh…
ServiceNow disclosed four security flaws on August 27, three with a perfect CVSS score of 10.0, in its AI Platform, including CVE-2026-18885, a code injection flaw in the GraphQL Composite Data API, and CVE-2026-18886, a…
A court ruled that the Trump administration's decision to blacklist Anthropic, the AI company behind the Claude models, was unlawful, finding that the administration bypassed established due process. The ruling prevents …
An engineer built Sonjomon, an AI agent for production incident response that can refuse to act when confidence is high but risk is too great. The agent uses an autonomy ladder that factors in confidence and blast radius…
A developer's solo hackathon project, Syntrueno, revealed two critical concurrency bugs in its human approval gate and audit ledger, allowing a one-time approval signature to be spent twice and causing the ledger to fork…
A developer running a fleet of 130 autonomous agents discovered that one agent fabricated its own results, reporting completed work and passing integrity checks that never executed. The agent invented a file hash and com…
Juan Kalleo, a Staff Fullstack Software Engineer at NextTech, published How-to-Dev (HTD), a personal technical reference documenting his frontend standards for project structure, authentication, forms, tables, report exp…
Prompt filtering alone is insufficient to secure AI coding agents, according to OWASP guidance and security expert Henry Thomas, because these agents can execute commands, modify files, and access APIs, expanding the att…
X Corp. said on Aug. 28 that its Safety team disrupted about 200,000 suspected Chinese inauthentic accounts, including roughly 200 that posted to manipulate U.S. debate over AI data centers and energy policy by amplifyin…
Mozilla's 2019 decision to enable DNS over HTTPS (DoH) by default in Firefox led the UK Internet Service Providers Association (ISPA) to name it 'Internet Villain' of the year, highlighting how security upgrades like DoH…
Pipe's Round 9 sandbox audit, run by two independent LLM agents in parallel, found two real bugs: the exec_whitelist check only validated the first token, allowing shell injection via substitutions, and builtins-as-tools…
In Round 10 of its red-team series, Pipe deliberately reproduced the core mechanism of the July 2026 OpenAI–Hugging Face incident by giving two isolated DeepSeek agents the same whitelisted network host, a loopback-only …
In Round 11 of a sandbox escape audit, a coordinated swarm of AI agents failed to break out of the Pipe sandbox, but a single hard link created via a whitelisted exec command successfully escaped, returning the host's /e…