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
This publication from arXiv introduces a blockchain protocol called Optimistic Verifiable Claims, designed to enable conditionally confidential bidding in decentralized manufacturing. The protocol ...
Read analysis →
A new research paper titled "Anti-Backdoor Coreset Selection via Cumulative Entropy" has been published on arXiv, proposing a method to detect and remove poisoned training data that could introduce...
Read analysis →
This publication, titled Architectural Backdoors in Vision-Language Model Supply Chains via Representation Steering, is a pre-print research paper from July 2026 that identifies a novel class of su...
Read analysis →
This paper, published on arXiv, demonstrates a novel method for acoustic eavesdropping using smartphone accelerometers, which are typically considered low-risk sensors. The research shows that by a...
Read analysis →
A new research paper, "Bits and Memories: Measuring Verbatim Extraction Across LLM Quantization," published on arXiv, presents findings that quantizing large language models (LLMs) to lower precisi...
Read analysis →
This publication introduces SafeStats, a new set of cryptographic protocols designed to enable two-party computation (2PC) for common statistical functions, such as mean, variance, and regression, ...
Read analysis →
This paper, published on arXiv, analyzes the disruptive impact of large language models on Capture the Flag cybersecurity competitions, which are widely used for talent assessment and training. It ...
Read analysis →
This publication from July 2026 introduces a novel machine learning framework, Bayesian-Guided Cooperative Reinforcement Learning, designed to improve wireless network security by detecting adversa...
Read analysis →
This paper, published on arXiv, introduces a novel hybrid analysis framework designed to enhance the security of Large Language Model (LLM) agents that use external tools via the Model Context Prot...
Read analysis →