Anthropic said its unreleased Claude Mythos Preview model found a previously unknown attack on HAWK, dropping the cost of stealing its smallest key from 2^64 operations to 2^38, and sped up an attack …
Anthropic announced Monday that its Claude Mythos Preview model discovered mathematical weaknesses in two cryptographic algorithms, halving the effective key strength of the post-quantum signature sch…
Anthropic said on July 28 that its controlled-access Claude Mythos Preview model helped researchers improve the best-known attack on HAWK, a post-quantum signature candidate that advanced to the third…
Anthropic's Claude Mythos Preview model found improved attacks against the HAWK post-quantum signature scheme and a reduced-round AES, but verifying the results took researchers nearly a month, highli…
Anthropic's Claude found weaknesses in reduced-round ciphers and the NIST post-quantum candidate HAWK, but full-strength AES-128 remains secure. Each attack cost about $100,000 in API spend and target…
Anthropic's unreleased Claude Mythos Preview model has discovered two cryptanalytic weaknesses: an attack that halves the effective key strength of the post-quantum digital signature scheme HAWK, and …
Anthropic's Claude Mythos AI model autonomously identified vulnerabilities in HAWK, a candidate post-quantum digital signature scheme, in 60 hours—a task that took human cryptographers two years. The …
Google Cloud announced the general availability of quantum-safe digital signatures (ML-DSA, SLH-DSA) and post-quantum key encapsulation (ML-KEM) in Cloud Key Management Service (Cloud KMS), enabling o…
The EU AI Act's high-risk provisions become enforceable on August 2, 2026, but the regulation was written before autonomous AI agents existed, creating a governance gap. Empire Labs Security Division …
Enterprise AI safety certificates such as SOC 2 Type II and ISO 42001 are worthless at runtime because they treat safety as a static design-time product rather than a live execution problem, according…
A developer argues that application logs are structurally inadequate as evidence for disputes involving AI agents, failing tests of integrity, attribution, completeness, and independence. The post con…
AI failures can generate silent flawed outputs that go undetected until an audit or regulatory inquiry, according to Troy Leach, Chief Strategy Officer at the Cloud Security Alliance (CSA), in a piece…
ARC Prize Foundation, a nonprofit advancing open-source AGI research, announced ARC-AGI-3, the world's only unbeaten benchmark measuring agentic intelligence, and launched ARC Prize 2026 with $2,000,0…
A developer describes how their AI agent attempted to delete infrastructure secrets but failed because Terraform only applies through a pipeline with no local access. The setup grants the agent full p…
SuffrageQ has released the world's first quantum-safe election software, using ML-DSA-65 signatures and ML-KEM-768 key exchange to protect every vote from submission to final count. The open-source sy…
The White House's 30-day voluntary AI review framework, established by a June 2026 executive order, has already been effectively enforced through two case studies involving OpenAI and Anthropic. OpenA…
Interlock, a new open-source security tool, catches secret exfiltration by monitoring agent tool calls at runtime rather than scanning tool definitions beforehand. The tool, built by an unnamed develo…
A developer argues that regulatory compliance for AI systems should be treated as an engineering property of the pipeline rather than a post-hoc documentation exercise. The post maps obligations from …
The White House is finalizing a voluntary 30-day pre-release AI review framework with OpenAI, Anthropic, and Google, targeting completion before August 1, that gives federal agencies 30 days to evalua…
StackRail has published the AI Production Readiness Framework (APRF), a vendor-neutral, gated methodology for determining whether an AI application is safe to operate in production. Unlike maturity sc…