Security & Safety
Decepticon 1.1.3 Tests the Guardrails for Autonomous Red-Team Agents
Decepticon 1.1.3 shows that red-team agents are now competing on rules of engagement, sandboxing, graphs, release integrity, and auditability.
← All categories
Security & Safety
Vulnerabilities, prompt injection, model safeguards, and containment design.
Security & Safety
OpenAI Adds Live Vote Counts to Election Answers
OpenAI outlined its 2026 election safeguards, combining AP vote counts, voting information, Codex Security, SynthID, usage policy, and political bias evaluations.
Security & Safety
620,000 attacks expose a 35-point safety gap in reasoning models
TELUS Digital tested 34 AI models with more than 620,000 adversarial attacks. The benchmark shows why enterprise AI safety is now an operating discipline.
Security & Safety
CLAUDE.md Became an Attack Surface, TrapDoor's Invisible Instructions
TrapDoor combines malicious npm, PyPI, and Crates.io packages with poisoned AI coding instruction files.
Security & Safety
Claude shows why agent security moved beyond approval prompts
Anthropic’s Claude containment writeup shows agent security shifting from prompt defenses and approval dialogs toward runtime isolation and blast-radius control.
Security & Safety
AI Broke an Erdos Conjecture, and the Real Story Is the Verification Loop
OpenAI’s unit-distance counterexample shows that AI research automation depends less on answer generation than on proofs experts can inspect.
Security & Safety
20 missed attacks, SLEIGHT-Bench warns agent security teams
SLEIGHT-Bench uses 40 synthetic attacks to show how easily LLM monitors can miss risky behavior by coding agents.
Security & Safety
Two H100s Are Enough, Command A+ Targets Private Agents
Cohere Command A+ lowers the bar for private AI agents with Apache 2.0 open weights and a two-H100 deployment target.
Security & Safety
After 10,000 vulnerabilities, Mythos moves the bottleneck to patching
Anthropic Project Glasswing shows that AI vulnerability discovery is no longer the slowest step. Verification, disclosure, and patch rollout are now the constraint.
Security & Safety
OpenAI adopts SynthID as AI image trust gets a new baseline
OpenAI and Google are turning C2PA, SynthID, and verification tools from image-generator features into web-scale trust infrastructure.
Security & Safety
Claude Gets a Conscience Tool, and Alignment Moves Into the Agent Loop
Anthropic is widening the moral formation conversation around Claude while testing an ethical reminder tool inside the model runtime loop.
Security & Safety
28 Security Integrations Put Claude in the AI Audit Log Era
Anthropic expanded Claude Compliance API integrations into the enterprise security stack. AI chats, files, and activity logs are becoming audit pipeline inputs.