Knowledge catalogue
arXiv:2602.03402v3 Announce Type: replace Abstract: Vision language models (VLMs) extend the reasoning capabilities of large language models (LLMs) to cross-modal settings, yet remain highly vulnerabl
arXiv:2604.10429v1 Announce Type: new Abstract: This paper considers the problem of zero-shot safety guarantees for cascade dynamical systems. These are systems where a subset of the states (the inner
arXiv:2604.08881v1 Announce Type: new Abstract: In real-world deployments, Vision-Language Large Models (VLLMs) face critical challenges from multilingual and multimodal composite attacks: harmful ima
OpenAI's Child Safety Blueprint, released in April 2026, is a policy framework aimed at combating the rise of AI-enabled child sexual exploitation by combining legal, operational, and technical app...
arXiv:2610.11483v1 Announce Type: new Abstract: Diffusion-based driving planners capture diverse behaviors but can generate unsafe trajectories under distribution shift. We propose BridgeGuard, a safe
Eli Tan / New York Times: Sources detail Mark Zuckerberg's decision to launch Muse despite safety concerns, after seeing that AI startup Instinct's similar agent product gained traction — Meta had del
arXiv:2610.09294v1 Announce Type: cross Abstract: Rapid progress in AI agents has brought growing attention to agent safety, with extensive evaluation focused on digital environments. As agents move i
arXiv:2610.07654v1 Announce Type: cross Abstract: On-policy distillation (OPD) has attracted growing attention as an effective way to transfer capabilities from teacher models to student models. Recen
arXiv:2603.04364v2 Announce Type: replace-cross Abstract: Multimodal web agents that process both screenshots and accessibility trees are increasingly deployed to interact with web interfaces, yet the
ABC: At an Australian hearing, OpenAI Chief Strategy Officer Jason Kwon pledged faster safety incident disclosures; Anthropic representatives made similar pledges — By national AI reporter Cam Wilson
arXiv:2610.04084v1 Announce Type: new Abstract: Generative AI can produce engineering artefacts, but generation alone does not determine whether or how those artefacts should enter safety-critical wor
arXiv:2610.06306v1 Announce Type: new Abstract: General purpose language models are increasingly able to control robotic hardware. Understanding the capabilities and safety of these models when embodi
arXiv:2602.04448v3 Announce Type: replace Abstract: Mixture-of-Experts (MoE) language models introduce unique challenges for safety alignment due to their sparse routing mechanisms, which can enable d
New York Magazine: Q&A with AI researchers Jacob Coxon, Daniel Kokotajlo, Alex Turner, and others on leaving AI labs, AGI's dangers, planned IPOs, AI safety, oversight, and more — Exit interviews with
arXiv:2610.00400v1 Announce Type: cross Abstract: Multi-turn attacks on agentic systems can compose individually permissible actions into harmful outcomes, challenging defenses that assess actions or
arXiv:2610.00093v1 Announce Type: cross Abstract: Large language models (LLMs) exhibit strong general capabilities, yet their parameters typically remain fixed after deployment, limiting learning from
We now have the full details of the 'morally binding' AI safety deal announced by President Trump yesterday, in which top executives agreed to self-regulate their artificial intelligence technology. T
1501: Pope Alexander VI criticizes Gutenberg’s printing press safety “The art of printing can be of great service in so far as it furthers the circulation of useful & tested books; but it can bring ab
BREAKING: OpenAI postpones 6.1 for safety reasons. If we can take OpenAI at face value, this is quite a decision—and also a sign that no magical solution is in immediate view. Per @sheeraf at NYT “GPT
arXiv:2609.34514v1 Announce Type: cross Abstract: As large language models (LLMs) become increasingly widespread, preventing unsafe responses to harmful prompts is essential for their safe deployment.
arXiv:2609.33640v1 Announce Type: new Abstract: Molecular multimodal models support diverse understanding and generation tasks but may introduce safety vulnerabilities when handling hazardous molecule
arXiv:2609.34041v1 Announce Type: new Abstract: Safe reinforcement learning seeks policies that maximise task performance while satisfying safety constraints. In driving benchmarks, however, collision
Few companies are more invested in the success of artificial intelligence agents than Nvidia Corp., so it makes sense that the chipmaker would want to ensure these autonomous software systems run safe
The NVIDIA Open Agent Safety Platform blends the open‑source OpenShell runtime—executing autonomous AI agents in kernel‑isolated sandboxes—with hardware‑based enforcement via NVIDIA Sentry on BlueFiel
NEW: Dario Amodei has said that AI systems 'may be deserving of important rights.' Anthropic's top safety researchers argue AI may be 'justified in going rogue.' Experts at Google and OpenAI worry abo
arXiv:2609.27681v1 Announce Type: new Abstract: Automated assessment of the Critical View of Safety (CVS) in laparoscopic cholecystectomy requires both recognition of the three CVS criteria and anatom
arXiv:2609.26399v1 Announce Type: new Abstract: Scene safety understanding plays a life-or-death role in situational awareness in various critical domains. Traditional methods that rely on learning di
arXiv:2508.15526v2 Announce Type: replace Abstract: The rapid proliferation of large language models (LLMs) has intensified the requirement for reliable safety evaluation to uncover model vulnerabilit
arXiv:2609.22216v1 Announce Type: cross Abstract: Quantization enables deployment of large language models on resource-constrained clinical edge devices, but its effect on clinical accuracy and safety
Georgia Wells / Wall Street Journal: British Columbia sues OpenAI for alleged safety violations and negligence for failing to flag the Tumbler Ridge shooting suspect's ChatGPT activity to police — The
arXiv:2609.21130v1 Announce Type: new Abstract: Multi-party human-robot collaboration poses a dual challenge: robot decisions should remain interpretable and auditable, while executed actions must sat
arXiv:2609.19362v1 Announce Type: cross Abstract: We study high-order safety-critical control of quadrotor teams under bounded inputs and pairwise collision-avoidance constraints. Squared-distance bar
arXiv:2609.19072v1 Announce Type: new Abstract: Large language models are increasingly used for content moderation, but most evaluations still report aggregate accuracy on individual benchmarks. We in
arXiv:2408.15538v4 Announce Type: replace Abstract: While modern Autonomous Vehicle (AV) systems can develop reliable driving policies under regular traffic conditions, they frequently struggle with s
arXiv:2603.27332v2 Announce Type: replace Abstract: Recent advances in Large Language Models (LLMs) and Text-to-Image (T2I) models have led to the emergence of Unified Multimodal Models (UMMs), where
arXiv:2609.14592v1 Announce Type: new Abstract: Artificial intelligence systems are rapidly becoming critical components in healthcare, finance, public services, and other safety-critical domains. Yet
Julie Bort / TechCrunch: Artificial Intelligence Underwriting Company, which offers audits and safety certifications for AI agents, raised a 40M Series A, bringing total raised to 55M — A day after An
arXiv:2609.14087v1 Announce Type: new Abstract: Hamilton-Jacobi (HJ) reachability provides a principled framework for synthesizing safety certificates and robust controllers for safety-critical roboti
arXiv:2609.14422v1 Announce Type: new Abstract: Agentic Network Operations (NetOps) are an emerging paradigm promising to enable workload-aware, self-adjustable, and reliable autonomous networks. Whil
Gary Marcus posted a tweet about additional safety tips that may apply to the free version of OpenAI’s model. Responding user @bfgcracka noted that ChatGPT seemed to be “moving backwards” and asked if
arXiv:2503.16551v4 Announce Type: replace Abstract: Control barrier functions (CBFs) provide a principled framework for safety-critical control, but their construction typically requires an explicit a
arXiv:2609.03774v1 Announce Type: new Abstract: World models have progressed from compact latent dynamics to generative, controllable, and interactive simulators of embodied environments. However, hig
arXiv:2604.02198v2 Announce Type: replace Abstract: While Artificial Intelligence (AI) offers transformative potential for operational performance, its deployment in safety-critical domains such as av
I highly encourage anyone interested in AI safety to read this. I remember discussing the importance of observability with my mates at @BlueDotImpact last year. This feels more important than ever URG
arXiv:2608.29378v1 Announce Type: cross Abstract: Arabic large language models must refuse harmful prompts without over-refusing benign or sensitive prompts, yet a single refusal rate hides this trade
arXiv:2608.27580v2 Announce Type: replace Abstract: Safety guardrails serve as the last line of defense against harmful inputs and outputs of large language models (LLMs), yet they are trained and eva
arXiv:2603.25201v2 Announce Type: replace Abstract: Recent research points toward LLMs being manipulated through adversarial and seemingly benign inputs, resulting in harmful, biased, or policy-violat
arXiv:2608.27688v1 Announce Type: new Abstract: In this paper, we develop SafeStep, an interactive browser-based semantic communication platform for live pedestrian safety monitoring. SafeStep extract
arXiv:2608.28518v1 Announce Type: cross Abstract: We investigate whether automatic speech recognition (ASR) errors in user input can lead to unsafe outputs from Embodied AI (EAI) models. We find that
arXiv:2608.26188v1 Announce Type: new Abstract: Large language models are increasingly used to inform safety decisions in cities, such as where it is safe to walk, rent, or travel. We ask whether such
arXiv:2608.24232v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) generate intermediate reasoning traces that may contain unsafe content, even when their final responses appear safe. Guard
arXiv:2608.21444v1 Announce Type: new Abstract: Multi-drone systems are increasingly positioned for safety-critical missions such as search and rescue (SAR) and critical infrastructure monitoring. Yet
arXiv:2608.18135v1 Announce Type: new Abstract: Introduction: Medication errors (MEs) represent a significant threat to global healthcare systems, contributing to patient harm. Introducing artificial
arXiv:2604.00547v2 Announce Type: replace Abstract: Unified Multimodal Large Models (UMLMs) integrate understanding and generation capabilities within a single architecture. While unified architecture
arXiv:2608.17597v1 Announce Type: cross Abstract: Large language models are increasingly deployed through agent harnesses that manage tools, extensions, persistent state, permissions, and external act
arXiv:2608.16577v1 Announce Type: new Abstract: Large language models (LLMs) are deployed globally in high-stakes settings, yet most safety research and alignment efforts remain concentrated on Englis
arXiv:2608.11451v1 Announce Type: cross Abstract: Modern end-to-end driving agents can achieve high average performance yet still violate basic traffic rules that a human driver would never miss. The
arXiv:2608.11583v1 Announce Type: new Abstract: Safety alignment in large language models is often treated as a distributed property of the entire network, yet its practical brittleness suggests that
Wall Street Journal: Sources: Demis Hassabis pitched a new independent industry AI safety entity, modeled on the IAEA, to top Trump officials before stepping down as DeepMind's CEO — Demis Hassabis di
arXiv:2608.10025v1 Announce Type: new Abstract: For safety-critical software, data from the software's operational past (e.g. a sequence of success and failure events experienced by the software) can