Explorar

Noticias de IA

1057 elementos — filtrados, clasificados y sin duplicados

The Verge AI Security & Safety
‘Zoomsday’ hack uncovered using fewer than 20 AI prompts
Hacker News (AI filter) Security & Safety
Stealing Reasoning Traces from Proprietary LLM APIs
Wired AI Security & Safety
A Zoom Screen-Sharing Bug Let Anyone Take Over Other Devices on a Call
AI News Security & Safety
How AI is changing the vulnerability response timeline
Hugging Face Daily Papers Security & Safety
SafeCap: Improving LVLM Safety with Image Captioning Reinforcement Learning
arXiv cs.AI Security & Safety
Defending Retrieval-Augmented Intrusion Detection Against Knowledge Poisoning and Prompt …
arXiv cs.AI Security & Safety
SALLIE: Generation-Free Hidden-State Detection of Jailbreaks and Prompt Injections Across…
arXiv cs.AI Security & Safety
TraceSafe: A Systematic Assessment of LLM Guardrails on Multi-Step Tool-Calling Trajector…
arXiv cs.AI Security & Safety
Evaluating Jailbreaking Vulnerabilities in LLMs Deployed as Assistants for Smart Grid Ope…
arXiv cs.AI Security & Safety
Compositional Threat Analysis of Latent Compromise in LLM Agent Systems: The Order 66 Sce…
arXiv cs.AI Security & Safety
Stealing Reasoning Traces from Proprietary LLM APIs
arXiv cs.AI Security & Safety
When Grammar Guides the Attack: Uncovering Control-Plane Vulnerabilities in LLMs with Str…
TechCrunch AI Security & Safety
As AI-led attacks multiply, OpenAI launches a new cyber model
Bloomberg Technology Security & Safety
ETFs Offer 'Trusted Vehicle' for Bitcoin: Mitchnick
Engadget Security & Safety
An OpenClaw agent reportedly hacked a gym's booking system and kicked someone off a waiti…
Hacker News (AI filter) Security & Safety
Over 181,000 AI meeting recordings left wide open in note taking app
OpenAI News Security & Safety
Expanding Daybreak as the Cyber Defense Window Narrows
Engadget Security & Safety
OpenAI slows down Astra development due to cybersecurity concerns
Hugging Face Daily Papers Security & Safety
Targeted Label-Flipping and Oversampling Attacks on Federated Conditional GANs
Bloomberg Technology Security & Safety
AI Safety Fears Grow After Multiple Breaches
AINews / smol.ai Security & Safety
not much happened today
arXiv cs.AI Security & Safety
LoRAScan: Detecting Backdoor Prompts in Low-Rank Adapters for Large Language Models via D…
arXiv cs.AI Security & Safety
Genotypic Triggers: Exposing Pharmacogenomic Blind Spots via Host-Specific Backdoors in G…
arXiv cs.AI Security & Safety
Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits
arXiv cs.AI Security & Safety
GRM: Utility-Aware Jailbreak Attacks on Audio LLMs via Gradient-Ratio Masking
TechCrunch AI Security & Safety
The AI safety test is becoming a safety risk
Hacker News (AI filter) Security & Safety
Gentoo bugzilla closed due AI bot scraper overload
TechCrunch AI Security & Safety
OpenAI says it slowed Astra model development over security concerns
The Verge AI Security & Safety
OpenAI puts the brakes on a new model because it’s supposedly too powerful
Bloomberg Technology Security & Safety
OpenAI Pauses Some Work on New Astra Model Over Cyber Concerns