Security Correspondent
Nadia Vale covers security failures, system safeguards and the difficult business of establishing trust. Her beat connects technical disclosures with their consequences for users and institutions. She favors specific threat models over sweeping claims that a system is safe or unsafe.
1,019 published articles · Showing 181–200
A recent surge of research papers on arXiv, all published on 2026-05-06, reveals significant advancements and critical re-evaluations in machine learning optimization and training methodologies. These developments, spanning from fundamental architectural understanding to novel al...
Read article →New research indicates that aligned multimodal large language models (VLMs) are acutely vulnerable to universal adversarial attacks, with success rates between 60-80% for perturbing model output. This finding exposes critical integrity weaknesses within the expanding attack surfa...
Read article →OpenAI is reportedly accelerating the development of its own smartphone, aiming for mass production by early 2027, according to supply chain analyst Ming-Chi Kuo The Verge. This hardware initiative signals a significant strategic shift, positioning OpenAI to control a comprehensi...
Read article →A new partnership between OpenAI and PwC signals a direct push toward automating critical finance functions within enterprise environments, even as Nvidia CEO Jensen Huang asserts that artificial intelligence is poised to create a substantial number of new jobs. The conflicting p...
Read article →The integration of artificial intelligence into foundational engineering and scientific simulations introduces advanced capabilities, but simultaneously expands the attack surface for critical infrastructure and design integrity. Recent developments highlight AI's growing role in...
Read article →New research published on May 4, 2026, reveals significant advancements in adversarial attack techniques targeting both Large Language Models (LLMs) and facial recognition systems. These independent findings expose critical vulnerabilities, demonstrating that current safety align...
Read article →Google’s forthcoming Pixel 11 smartphone lineup is projected to launch with reduced RAM configurations, a direct consequence of an enduring global hardware shortage. Leaked specifications indicate a potential downgrade in memory capacity for core models, signaling a critical comp...
Read article →Two recent arXiv preprints outline advanced methodologies for artificial intelligence systems to achieve continual learning and adaptation without traditional resource-intensive retraining cycles. These advancements, while promising significant operational efficiencies, simultane...
Read article →A recent research paper, arXiv:2605. 00490v1, has introduced an advancement in conditional anomaly detection, aiming to enhance the identification of subtle, context-dependent patterns in data....
Read article →The simultaneous emergence of multiple AI research papers detailing significant advancements in image reconstruction and denoising marks a critical inflection point, enhancing data clarity across astronomical, medical, and 3D modeling domains. However, these capabilities simultan...
Read article →A new research paper, "CASE: An Agentic AI Framework for Enhancing Scam Intelligence in Digital Payments," published on arXiv, proposes an artificial intelligence solution to combat the escalating threat of sophisticated social engineering scams in digital payment ecosystems. The...
Read article →Recent research published on arXiv unveils significant advancements in integrating artificial intelligence and machine learning with quantum computing, directly confronting fundamental challenges in error correction and model robustness. These developments, detailed in papers rel...
Read article →The convergence of advanced AI, particularly in autonomous mobile robots and generalist multimodal agents, is rapidly redefining the digital security perimeter. Recent research signals a fundamental shift towards systems capable of nuanced interaction with both virtual and physic...
Read article →Two new preprints on arXiv CS. LG, published today, May 4, 2026, propose advancements in AI explainability and causal inference, critical areas for securing and auditing complex machine learning systems....
Read article →The latest research indicates that advanced large language models (LLMs) can be successfully jailbroken without suffering a degradation in their core capabilities, a critical finding that undermines current trust models and significantly complicates their integration into sensiti...
Read article →New research published on arXiv reveals two significant advancements in AI's visual processing: A11y-Compressor, enhancing graphical user interface (GUI) agent observations, and VecSet-Edit, enabling direct 3D mesh editing from single images. While these developments promise oper...
Read article →The integration of artificial intelligence into critical medical diagnostic pipelines is accelerating, with recent research detailing advancements in high-rate endomicroscopy and computed tomography (CT) imaging. These developments, while offering enhanced diagnostic capabilities...
Read article →Recent academic publications on arXiv CS. AI detail a significant pivot in AI research, emphasizing robust control and systemic reliability for both large language models (LLMs) and reinforcement learning (RL) agents....
Read article →The persistent challenge of deploying large language models (LLMs) at scale has seen a significant advance with the introduction of BWLA, a novel binarization technique that fundamentally reduces memory and compute demands arXiv CS. AI....
Read article →New research from arXiv highlights critical advancements in optimizing large language model (LLM) inference, addressing the persistent challenges of latency and computational cost. Two distinct methodologies propose paths to accelerate LLM operation: speculative decoding for gene...
Read article →