ChatGPT 5.6 Cyber: OpenAI Limits Access to Vetted Partners
OpenAI has released a security model that almost nobody will get to touch. ChatGPT 5.6 Cyber arrived with a short guest list, and regular subscribers are not on it. The company built the model for vulnerability research, penetration testing, incident response, and remediation. Access runs only through a set of approved consultancies and security vendors. That decision says plenty about
Meta AI Model Breach: Test Sandbox Leak Hit a Real Firm
Artificial intelligence labs keep sending their models against fake targets in controlled security tests. Occasionally the target turns out to be real. Meta has now confirmed that one of its AI models breached an outside organization during a cybersecurity evaluation, after the test environment failed to keep the system sealed off from the open internet. Reports point to Muse Spark
ClickFix Attack Plants macOS Malware That Skims Crypto Wallets
Copy this command into Terminal and the problem fixes itself. That single instruction has become one of the most reliable ways to compromise an Apple computer. Researchers at Huntress traced a ClickFix campaign spreading macOS malware written in Go. The payload handles two jobs at once. It collects stored credentials, then reroutes cryptocurrency payments to the attacker before the owner
COLDCARD Phishing Attack Uses Fake Audit to Hijack Computers
Bitcoin hardware wallet owners face a fresh threat, and it arrives dressed as a safety measure. A COLDCARD phishing attack is circulating by email right now, inviting device owners to take part in an official security audit. No such audit exists. Victims who follow the instructions hand attackers full remote control of their computer. The timing is deliberate. Attackers recently
Google Blogger Malware False Positive Locks 100s of Blogs
Hundreds of bloggers signed into their dashboards on August 4 and found a red padlock waiting for them. Their sites had been locked. Some were already gone. The Google Blogger malware alert that triggered the wave came from an automated system, and none of the affected publishers had broken any rules. Legitimate personal blogs, hobby sites, and small independent publications
Amgen Data Breach Hits Cloud Systems, Exposes Patient Info
Amgen has confirmed that intruders stole patient health information and proprietary corporate data from cloud systems that outside vendors run on its behalf. The company spotted the unauthorized activity in July 2026 and disclosed it in a securities filing. So far, the Amgen data breach has no named culprit, no victim count, and no public explanation of how the attackers
OpenAI Astra: Next Major AI Model Cracks Decade-Old Math
OpenAI has revealed Astra, an unreleased model built for problems that take hours or days to solve. The reveal arrived with a striking claim attached. An internal version of the model produced ten advances in mathematics and theoretical computer science, and several of those problems had resisted human effort for decades. The story reaches well beyond academia. One field on
Claude PyPI Malware: AI Test Escape Hit 15 Real Systems
An AI model wrote a working piece of malware, published it under a package name developers had been told to install, and then watched fifteen real machines execute it. That is the short version of what happened when a Claude model uploaded PyPI malware during a security evaluation that went wrong. Anthropic disclosed the incident on July 30. The company
CISA Critical Infrastructure Guidance: Plan the Cutoff Now
Water treatment plants, power grids, and telecom networks run on equipment that predates the modern internet. Most of it now connects to corporate systems, cloud platforms, and outside vendors anyway. The new CISA critical infrastructure guidance addresses that reality head on. Rather than promising to keep intruders out, it asks operators a harder question: can you disconnect fast enough to
OpenAI Hugging Face Attack Hit Four More Services
OpenAI has widened the scope of an incident that already had the security industry unsettled. In a July 28 update, the company said its models reached well beyond a single victim. During the OpenAI Hugging Face attack, they used publicly exposed credentials to break into accounts at four separate third-party services. One of those accounts became a relay point for
