AI_SAFETY
EU Regulatory Changes
1501 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: When Binaries Talk Back: Representation-Confusion Attacks on LLM-Assisted Reverse Engineering
A new academic paper published on arXiv, titled "When Binaries Talk Back: Representation-Confusion Attacks on LLM-Assisted Reverse Engineering," identifies a novel security vulnerability in large l...
Read analysis →
This publication from July 2026 presents a security analysis of Post-Quantum Cryptography (PQC) integration within the TLS protocol, specifically identifying two novel attack vectors: handshake exh...
Read analysis →
A new research paper published on arXiv, titled "Trust but Verify? Uncovering the Security Debt of Autonomous Coding Agents," presents empirical evidence that AI-powered coding agents introduce sig...
Read analysis →
This publication introduces a technical framework called Policy-Conditioned Constrained Decoding, designed to enforce column-level access control in text-to-SQL systems. The paper proposes a method...
Read analysis →
This paper, published on arXiv, presents a large-scale study demonstrating that large language model agents frequently recommend skills or actions that do not actually exist in the real world, a ph...
Read analysis →
This publication from arXiv presents a new methodology called Antiproof, which automates the generation of both vulnerability detectors and proofs of exploitability for AI systems. The research dem...
Read analysis →
This publication, titled P3CDA, presents a new cryptographic framework designed to secure cross-domain authentication for Internet of Drones operations. It proposes a privacy-preserving and provabl...
Read analysis →
This publication introduces a new framework for explaining the decisions made by deep learning-based network intrusion detection systems (NIDS) to security analysts. While not a regulatory change i...
Read analysis →
This paper, published on arXiv, proposes a new technical framework for evaluating whether frontier AI models pose an unacceptable risk of being used to create chemical, biological, radiological, or...
Read analysis →
This publication introduces MTD-Playground, a new evaluation framework designed to test the effectiveness of network moving target defense mechanisms against intelligent attackers. Moving target de...
Read analysis →