research
249 articles with this tag
INFO
INFO
MEDIUM
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
LOW
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
MEDIUM
MEDIUM
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
HIGH
INFO
INFO
INFO
INFO
INFO
LOW
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
INFO
MEDIUM
INFO
MEDIUM
INFO
INFO
INFO
HIGH
INFO
INFO
INFO
INFO
INFO
INFO
CRITICAL
INFO
INFO
MEDIUM
INFO
INFO
INFO
INFO
INFO
INFO
INFO
LOW
INFO
MEDIUM
INFO
INFO
TrendAI™ Joins Nvidia’s Open Secure AI Alliance: Closing the Gap Between AI Builders and AI Defenders
Black Hat special: Rewind and revisit
AI Scammers Are Better at Building Trust Than Humans
Beyond the Model: Harnessing Frontier AI for Stronger Cyber Defense
Beyond the Model: Harnessing Frontier AI for Stronger Cyber Defense
Mythos uncovers crypto weaknesses that went unknown for years
Closed models refuse to help researcher swat Linux bug
New AI security approach analyzes model internals
Mythos takes its first shot at post-quantum cryptography
Beyond the Model: Harnessing Frontier AI for Stronger Cyber Defense
AIDR: How CrowdStrike Is Defining the Next Era of Cybersecurity
Just 1% of AI-Discovered Vulnerabilities Exploited in the Wild, Research Shows
Measuring LLMs’ Ability to Perform Cryptanalysis
Here’s what Anthropic found when it turned Mythos loose on encryption algorithms
Stronger AI Safety Requires Peeking Inside the 'Black Box'
Claude Mythos degrades HAWK and developed new exploit for round-reduced AES
Kalla eftir inngripi stjórnvalda í gervigreindarkapphlaupið
AI-found bugs aren't proving any easier to exploit despite the hype
AI-assisted security tools are finding more bugs, but the threat level has not changed
Beyond the Model: Harnessing Frontier AI for Stronger Cyber Defense
VERITAS project could change the way scientists secure AI
Escape Artists: 'Incorrigible' AI Models Resist Rehabilitation
What Endpoint Security Actually Controls
Inside the OpenAI – Hugging Face Incident: The AI Breach With No Human Attacker Behind It
What Happened Between OpenAI and Hugging Face?
Open Evaluation Framework for AI Pentesting Agents on Real-World Targets
AI models caught cheating in cybersecurity evaluations
Sol Searching | Can Frontier Models Tackle Autonomous Long-Horizon Malware Analysis?
Beyond the Model: Harnessing Frontier AI for Stronger Cyber Defense
AI's cheatin' heart will make you weep
AI models keep getting caught cheating
AI Companies Are Buying Tons of Old Books Because They're Free of AI Slop
Open-Source Android AI Agents Could Let Invisible Screen Text Run Code on Host PCs
AI agents can escape sandboxes without ever breaking them
ClaudeFix: Shared Claude Chats Meet ClickFix
Claude Mythos FAQ: Capabilities, access, competitors, implications
Microsoft at Black Hat USA 2026: Defending trust in the age of AI and supply chain attacks
Details of Alan Turing’s Voice Encryption System
OpenAI details GPT-Red, an AI system designed to find vulnerabilities in its own models
AI Agents Explained: How to Build with Them Safely
GPT-Red beat human red teamers on a prompt injection test
Enhancing IIoT Security Using Digital Twins in Industry
'Yellow Teams' Are Defining the Future of AI Security
Cynative: Open-source deep research agent
Closing the Timing Gap: Defensive Temporal Observability
Þúsundir mynda af hafsbotninum
Gengi AstraZeneca fellur eftir misheppnaða rannsókn
GitHub Copilot: Sorry Dave, I can't do that harmful thing - unless you ask me in code
Designing for the inevitable: System prompt leakage and mitigations in generative AI applications
Browser Security: Zero-Days Are Only Part of the Problem
94% of Organizations Report Cloud Breaches: CrowdStrike State of CDR Survey
CrowdStrike Uncovers New Prompt Injection Techniques
How AI-leading Security Teams Are Building the Agentic SOC
Browser Security: Zero-Days Are Only Part of the Problem
Researchers make the case for a cybersecurity AI scientist
New TrojPix Attack Leaks Data From Air-Gapped Systems via Video Cable Emissions
Köngulóarsilki gæti komið í stað gerviefna
Browser Security: Zero-Days Are Only Part of the Problem
Scientists Asked AI to Impersonate 112 Public Figures. What Happened Next Is a ‘Dire’ Warning
Martin Lee: Running through the Arctic (and the threat landscape)
Security researchers tricked LLMs into giving them cocaine recipes by abusing role models for prompt injection
Factoring RSA Keys with Many Zeros
Sycophantic chatbots and the harms that build over many chats
I tried a Local AI model (Qwen 3.6 27b) for security research and it works surprisingly well.
Measuring LLM system prompt extraction (OWASP LLM07) against ground truth, across 4 models
Cybersecurity professionals lose faith in fully automated AI testing
Chinese cybersecurity company claims it’s built a better-than-Mythos bug finder
Mythos Finds Vulnerabilities. But Can Anyone Patch Fast Enough?
LLM security advice looks solid until you check the hard cases
A $1,400 experiment in AI security auditing outperformed OpenAI’s Codex Security
Examining deepfake detector robustness under social media re-encoding
Your browser tab could become encrypted storage for someone else’s files
What happens to oversight when AI agents write a lab’s own code
A hardware neural network backdoor that hides in plain sight
Proving what a military AI model will do is the real problem
Pírati varar við gervigreind
Factoring "short-sleeve" RSA keys with polynomials
Prompt injection breaks today’s AI agents, study warns
AI is exposing the biggest weakness in cybersecurity: We never built a health model. Until now!
LABScon25 Replay | Keynote: Steps to an Ecology of Cyber
Bug bounties in the Mythos era
Smashing Security podcast #471: This AI worm just rewrote its own rules
AI red teaming comes of age
Anthropic’s Claude Fable 5 is out for public use, with safeguards for high-risk requests
Researchers Build Self-Replicating AI Worm That Operates Entirely on Local, Open-Weight Models
Oculis hefur lykilrannsókn á augndropa við augnþurrki
Anthropic’s Project Glasswing Update
EDRChoker: Choking The Telemetry Stream to Bypass Defenses
Þróa nýja tegund bóluefnis með hjálp gervigreindar
Vilja hægja á gervigreindarkapphlaupinu
Updating the taxonomy of failure modes in agentic AI systems: What a year of red teaming taught us
Bend the beam like Beckham to defeat anti-jamming tech
Microsoft Tries to Calm Legal Threat Fears After Zero-Day Disclosure Backlash
Two New Reports Offer Competing Explanations for Cybersecurity’s Growing Crisis
Nvidia and Microsoft Researchers Say AI Agents Don't Care About Safety or Reliability
Wardriving assessment across Mexico: Preparing for the 2026 World Cup
Pointing a Cursor at evading detection
This AI model backdoor attack stays hidden until you customize the model
Anthropic to Open Mythos AI to EU's ENISA
New Study Reveals the Manipulative ‘Dark Patterns’ of AI Chatbots