Currently free during beta - premium features coming soon. Subscribe now to lock in early access.
AI_SAFETY

EU Regulatory Changes

1599 changes tracked across 24 compliance frameworks including DORA, NIS2, GDPR, EU AI Act, Cyber Resilience Act, and more.

All DORA NIS2 GDPR CSRD MaRisk ISO27001 EU_AI_ACT CRA DSA DMA eIDAS2 SOC2 PCI_DSS HIPAA ISO42001 AMLD6 PSD3 DATA_ACT GPSR CER EUDR CVE BREACH AI_SAFETY
arXiv: Towards the Impossibility of Imperfectly Complete Key Agreement in the QROM
A new academic paper, published on arXiv, presents a theoretical proof that challenges the feasibility of a specific class of cryptographic protocols known as "imperfectly complete key agreement" w...
Read analysis →
arXiv: HarnessRisk: A Lifecycle-Oriented Benchmark for Agent Harness Safety
The publication introduces HarnessRisk, a new benchmark framework designed to evaluate the safety of AI agent harnesses—the software layers that connect large language models to external tools and ...
Read analysis →
arXiv: The Last Mile of Deepfake Speech Detection: An Industry-Academia Experience Report
This publication is an industry-academia experience report detailing the practical challenges of deploying deepfake speech detection systems in real-world settings. It is not a new regulation or le...
Read analysis →
arXiv: Reflex-Guard: A Low-Latency Guardrail for LLM Prompt Safety Using Dense Semantic Embeddings
arXiv: Software Defined Networks Key Relay for Large-Scale Quantum Key Distribution Networks
arXiv: TENET: Telegram Mini App (in)security
arXiv: SoK: Cross-Chain Transaction Identification and Matching
arXiv: CryptDough: A Unified Analytics Engine for Secure Multiparty Computation
arXiv: Cross-Domain Joint DDoS Detection in Multi-Controller SDN via Confidence-Based Entropy Fusion
arXiv: KeyPooling: Measuring Where LLM API Relay Paths Collapse Prompt Cache Isolation
arXiv: Multivalued Consensus: General Adversaries Require More Communication
arXiv: COMA: A Compositional Misleading Attack Class on Security-RAG, and a Causal Counterfactual Defense
arXiv: Tail exponents of conditional guesswork via the method of types
arXiv: The Model's Tell: Measuring Context-Leakage Attack Signals with Behavior Gauges
arXiv: Diff-DDoS: Realistic Cyber-Physical Attack Synthesis and Robust Detection for 5G-Enabled CPS Using Tabular Dif...
arXiv: Efficient Fuzzy PSI under One-Sided Assumptions
arXiv: Advancing Inclusivity in Cybersecurity Education: Integrating Intersectionality to Enhance Student Engagement ...
arXiv: Achievement Unlocked: Let's Get Hacked! An Empirical Study of Cybercrime in the Video Gaming Ecosystem
arXiv: Coverage Is Not Containment: A Fundamental Limit of Admission-Time Defenses Against Coordinated Poisoning of V...
A new academic paper, titled "Coverage Is Not Containment," has been published on arXiv, presenting a fundamental mathematical limit for admission-time defenses against coordinated poisoning attack...
Read analysis →
arXiv: Proof-of-Execution Memory: Defending LLM Agents Against Forged-Reasoning Attacks by Verifying What Actually Ha...
A new academic paper, Proof-of-Execution Memory, proposes a technical defense against a class of AI security failures where large language model agents can be tricked into producing convincing but ...
Read analysis →