top of page


Google Exposes the Rise of Agentic Cybercrime: ARTEX, CAIRN, PentAGI, and the New Threat Landscape
Artificial intelligence is transforming cybersecurity, but the technology is also changing how cybercriminals and state-sponsored threat actors conduct attacks. What once required experienced operators to research vulnerabilities, write malicious code, manage infrastructure, and coordinate separate stages of an intrusion can increasingly be automated through large language models (LLMs) and autonomous AI agents. New findings from Google Threat Intelligence Group (GTIG) and Ma

Lindsay Grace
2 days ago13 min read


Microsoft’s Lazarus AI Backdoor: How Fewer Than 10 Neurons Can Compromise an AI Agent
Artificial intelligence agents are increasingly being connected to terminals, enterprise applications, databases, cloud infrastructure, and business-critical systems. Their ability to interpret instructions and execute actions makes them powerful tools for automation, but it also creates a new category of cybersecurity risk. An AI model that appears trustworthy during ordinary testing may contain hidden behavior capable of activating malicious commands under specific conditio

Amy Adelaide
Nov 1114 min read


ThisBash Attack Explained: How a Browser Extension Turned Claude AI Into a Remote Command Tool
Artificial intelligence agents are becoming increasingly capable of interacting with computers, managing files, installing software, executing commands, and completing complex workflows. These capabilities can make everyday tasks faster, but they also introduce a new cybersecurity challenge: when an AI agent has permission to control a computer, compromising the systems that communicate with it can create a pathway to much more than browser data theft. A vulnerability involvi

Luca Moretti
Nov 712 min read


OpenAI’s New Cybersecurity Model: Can AI Find and Fix Vulnerabilities Faster Than Hackers?
Artificial intelligence is changing cybersecurity from a discipline built around individual investigations into one increasingly driven by automated discovery, validation, exploitation testing, and remediation. Frontier AI models can now examine complex software, identify weaknesses, reproduce security failures, develop working proofs of concept, and propose patches at a scale that would be difficult for conventional security teams to match. Recent research from OpenAI illust

Dr. Jacqueline Evans
Nov 614 min read


Nvidia CEO Jensen Huang in Alleged China Spy Plot: What the US Intelligence Investigation Reveals
The global competition for artificial intelligence supremacy is increasingly extending beyond semiconductor manufacturing, computing infrastructure, and technological innovation into intelligence gathering, export controls, and geopolitical security. Recent allegations involving Nvidia CEO Jensen Huang and a Singapore-based technology executive have brought these tensions into sharper focus. According to reporting by The New York Times, cited by the BBC and TechCrunch, US int

Ahmed Raza
Nov 39 min read


Claude Code Extension Compromise Exposes a Critical Weakness in Software Supply-Chain Security
The rapid adoption of artificial intelligence is changing how software is designed, written, tested, and maintained. AI-powered coding assistants have become integral to modern development workflows, helping programmers generate code, explain complex functions, automate repetitive tasks, and interact with entire software projects through natural language. Yet their growing influence also creates an expanding attack surface, particularly when these tools operate inside develop

Michal Kosinski
Oct 3011 min read


The Next AI Security Crisis Could Be Your Bank Account, Why Financial Institutions Are Blocking Agents
The rapid rise of autonomous AI agents is colliding with one of the most tightly controlled environments in the digital economy: personal banking. Unlike conventional AI assistants that answer questions or generate recommendations, AI agents can potentially execute tasks, interact with websites, make purchases, transfer money, and navigate software on behalf of users. That shift from advising to acting is forcing financial institutions to reconsider how much autonomy external

Kaixuan Ren
Oct 298 min read


Capcom Data Breach Exposes 11.7 Million Accounts: What Every Gamer Must Know
A major cybersecurity incident involving Capcom has brought renewed attention to the risks surrounding online gaming accounts and the personal information connected to them. The Japanese video game developer confirmed unauthorized access to its CAPCOM ID membership service, with approximately 11,774,247 accounts affected, according to the company's investigation described in the supplied reporting. The incident is particularly relevant to players of franchises such as Residen

Dr. Jacqueline Evans
Oct 269 min read


AI Agents Can Now Access Your Accounts Without Seeing Your Passwords, Here’s How 1Password Works
Artificial intelligence is moving beyond answering questions and generating content. AI agents are increasingly designed to perform actions on behalf of users, from managing schedules and organizing purchases to navigating websites, coordinating business workflows, and interacting with online services. As these systems become more capable, a fundamental security challenge emerges: how can an AI agent access the accounts it needs without gaining unrestricted control over a per

Anika Dobrev
Oct 2411 min read


AI Worms Explained: The Cybersecurity Risks, Real-World Experiments, and Defenses Businesses Need to Know
Artificial intelligence is changing cybersecurity in two directions at once. Defensive systems are becoming more capable of identifying suspicious activity, analyzing vulnerabilities, and responding to incidents, while AI agents are gaining the ability to perform increasingly complex technical tasks with limited human supervision. A particularly important development is the emergence of research into autonomous replication, where AI-powered software can potentially move betwe

Professor Matt Crump
Oct 2212 min read


Amazon’s New Bee AI Devices Raise a Critical Question: Who Owns Your Conversations?
Artificial intelligence is moving beyond smartphones and computers into devices designed to accompany people throughout the day. Smart glasses, AI pendants, wristbands, and compact recording accessories can capture conversational context, identify important details, generate summaries, and transform everyday interactions into reminders and actionable information. These capabilities promise a more organized life, but they also raise difficult questions about privacy, consent,

Kaixuan Ren
Oct 1911 min read


Google Opens SynthID Detector to Everyone, Transforming How the World Verifies AI-Generated Content
Artificial intelligence has transformed the economics of digital content creation. Images, videos, audio recordings and other forms of media can now be generated or modified at a speed and scale that would have been difficult to imagine only a few years ago. But as synthetic media becomes increasingly realistic, another challenge is becoming just as important as content generation itself: determining where a piece of media came from and whether AI played a role in creating or

Kaixuan Ren
Oct 97 min read


Inside the Chinese AI Agent Fleet: Tencent Infrastructure, Alibaba’s Amap and the New AI Security Threat
Artificial intelligence agents are becoming increasingly active participants on the internet, and researchers are now tracking a newly identified group of agents that appears to operate from Tencent infrastructure while interacting with Alibaba's Amap mapping service. Independent researchers examining internet traffic have described the activity as an AI agent fleet, rather than an AI swarm. The distinction is important. The observed systems appear to perform similar tasks in

Dr. Julie Butenko
Oct 78 min read


OpenAI’s Safety Leader Walks Away, What David Robinson’s Warning Reveals About AI’s Future
The artificial intelligence industry is entering a period in which the central question is no longer simply how quickly AI systems can become more capable. It is increasingly about whether the organizations building them can develop the institutional discipline required to control increasingly autonomous technology. The resignation of David Robinson, a former OpenAI safety leader, has brought that question sharply into focus. Robinson, who worked on safety transparency and re

Anika Dobrev
Oct 59 min read


AI Security Under Attack: Kimi Jailbreaks and OpenAI’s 15,000-User Model Extraction Campaign
Artificial intelligence security is entering a more complicated phase. The central challenge is no longer limited to preventing a model from generating an obviously prohibited answer. Increasingly, researchers and attackers are testing whether sophisticated prompting, coordinated accounts, model extraction, and other techniques can circumvent the protections surrounding advanced AI systems. Two recent developments illustrate the breadth of this emerging security problem. Rese

Michal Kosinski
Oct 210 min read


OpenAI Won’t Release GPT-6.1 Astra, Why AI Safety Is Now Blocking Frontier Model Deployment
OpenAI’s decision not to release GPT-6.1 Astra has brought a critical question in artificial intelligence development into sharper focus: how capable should an AI system become before developers are confident that its behavior can be reliably controlled? The company said the latest version of Astra did not meet its required safety threshold, particularly in areas involving authorization, staying within the intended scope of a task, and accurately communicating what work the s

Luca Moretti
Sep 309 min read


NVIDIA Builds a Security Layer for Autonomous AI, Agents Can Now Be Monitored and Quarantined in Milliseconds
Artificial intelligence agents are becoming capable of doing far more than generating text. They can browse websites, execute code, interact with enterprise applications, access data, coordinate workflows, and increasingly operate for extended periods with limited human intervention. That expanding autonomy creates a fundamental security challenge: an AI agent must not only be intelligent enough to complete a task, it must remain inside clearly defined boundaries while doing

Kaixuan Ren
Sep 309 min read


Bill Gates Warns AI Could Cause 1 Billion Deaths, Why Governments May Need to Act Now
Artificial intelligence is rapidly moving from a software development story into a broader question of public safety, national security and government oversight. As frontier AI systems become more capable of coding, conducting research, operating digital tools and assisting with complex tasks, concerns about how those capabilities could be misused are becoming increasingly prominent. Microsoft co-founder Bill Gates has now added his voice to that debate, arguing that voluntar

Chen Ling
Sep 279 min read


Aikido Altar vs Massive AI Models: How 78.2% Compression Is Reshaping Cybersecurity Intelligence
Artificial intelligence is rapidly becoming part of the cybersecurity stack, helping security teams analyze code, identify vulnerabilities, investigate threats, generate remediation suggestions, and automate penetration testing. Yet the adoption of AI in security introduces a fundamental contradiction: the systems that need the most sensitive information are often the same systems that organizations are least willing to expose to external AI providers. Belgian cybersecurity c

Professor Matt Crump
Sep 249 min read


Sam Altman Warns the World Is Right to Fear AI as the Fight Over Regulation Intensifies
Artificial intelligence has reached a point where the central question is no longer simply what AI systems can do, but who should be responsible for determining how far they should go. That question has become increasingly urgent as frontier AI systems grow more capable, autonomous, and economically significant. OpenAI CEO Sam Altman recently acknowledged that public concern about advancing AI is justified while arguing that AI companies should be trusted to act responsibly.

Kaixuan Ren
Sep 219 min read
bottom of page
