Tag: control
-
US Lifts Export Curbs on Anthropic AI Models
Commerce Ends 18-Day Ban to Restore Global Access to Fable 5, Mythos 5. The U.S. Department of Commerce lifted export controls on Anthropic’s Fable 5 and Mythos 5, ending an 18-day restriction imposed over national security concerns. The reversal restores global access. It is unclear whether the government will use similar controls on future AI…
-
U.S. Commerce Withdraws Export Controls on Anthropic Claude Models After Security Commitments
The U.S. Department of Commerce has recently lifted export controls on Anthropic’s advanced AI models, Claude Fable 5 and Mythos 5, following a series of security and compliance commitments made by the company. This decision represents a significant shift in the regulatory landscape surrounding frontier artificial intelligence systems, particularly those that could be used for…
-
Microsoft wants to stop unwanted bots from entering Teams meetings
A new Microsoft Teams admin policy, Manage external bots and their access to meetings, gives organizations greater visibility and control over external bots in meetings. The … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/07/01/microsoft-teams-bot-detection-and-protection/
-
Anthropic to restore Claude Fable access on Wednesday
Anthropic has confirmed that the Department of Commerce has lifted export controls on Claude’s two most powerful models, Fable 5 and Mythos 5. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/anthropic-to-restore-claude-fable-access-on-wednesday/
-
Malicious PyPI packages give hackers control of Telegram bot servers
A campaign active since last November has been targeting Python developers building Telegram bots with trojanized Pyrogram forks that allow attackers to read arbitrary files on compromised servers. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/security/malicious-pypi-packages-give-hackers-control-of-telegram-bot-servers/
-
Austria Urges Anthropic to Move to EU to Avoid US Controls
Mythos and Fable Export Controls Deprive EU of ‘Cutting-Edge Innovation,’ Security. Stung by the Trump administration’s export controls on Anthropic’s most powerful cyber-capable models, Mythos and Fable, the Austrian government wants Europe to tempt Anthropic into moving across the Atlantic. Austria told the EU sovereignty leader that we must act now. First seen on govinfosecurity.com…
-
Cloudflare und beehiiv integrieren AI Crawl Control für unabhängige Publisher
AI Crawl Control wird laut Ankündigung in der Beta-Phase allen beehiiv-Nutzern Einblick geben, wie KI-Dienste mit ihren Inhalten interagieren und welchen Traffic sie erzeugen. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/cloudflare-und-beehiiv-integrieren-ai-crawl-control-fuer-unabhaengige-publisher/a45631/
-
Millenium RAT Uses Base64 and XOR Configuration to Hide Telegram C2 Settings
Millenium RAT version 4.* exposes a compact but potent evolution: the malware has migrated from .NET to native C++, while retaining a stealthy Telegram-based command-and-control (C2) model that requires no bespoke server infrastructure. The sample set and telemetry analyzed by Group-IB show the RAT embeds its entire configuration inside an RCDATA resource, masks that configuration…
-
China’s Zhipu AI Model GLM-5.2 Detects Software Vulnerabilities Like Claude Mythos
Zhipu AI’s newly released GLM-5.2 model is attracting significant attention from the cybersecurity community due to its vulnerability detection capabilities, which are comparable to those of Anthropic’s restricted Claude Mythos system. This development raises new concerns about the effectiveness of U.S. export control policies on advanced artificial intelligence. Released on June 13, 2026, under a…
-
Critical Linux Kernel Flaw Allows Unprivileged Users to Gain Full Root Access
A newly disclosed flaw in the Linux kernel’s traffic-control subsystem, now assigned CVE-2026-46331 and referred to as >>Pedit COW,<< has been found to grant any unprivileged local user full root access on vulnerable systems. Within just 24 hours of the CVE being formally assigned on June 16, 2026, a working proof-of-concept exploit dubbed packet_edit_meme surfaced…
-
Post-Quantum Security Spurs National Sovereignty Thinking
AI Export Controls Expose Hidden Risks to Post-Quantum Cryptography Migrations. Security leaders warn that post-quantum cryptography migration is creating new dependencies on foreign vendors, hyperscalers and supply chains, raising questions about resilience, crypto-agility and national control over critical security infrastructure. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/post-quantum-security-spurs-national-sovereignty-thinking-a-32095
-
Post-Quantum Security Spurs National Sovereignty Thinking
AI Export Controls Expose Hidden Risks to Post-Quantum Cryptography Migrations. Security leaders warn that post-quantum cryptography migration is creating new dependencies on foreign vendors, hyperscalers and supply chains, raising questions about resilience, crypto-agility and national control over critical security infrastructure. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/post-quantum-security-spurs-national-sovereignty-thinking-a-32095
-
Post-Quantum Security Spurs National Sovereignty Thinking
AI Export Controls Expose Hidden Risks to Post-Quantum Cryptography Migrations. Security leaders warn that post-quantum cryptography migration is creating new dependencies on foreign vendors, hyperscalers and supply chains, raising questions about resilience, crypto-agility and national control over critical security infrastructure. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/post-quantum-security-spurs-national-sovereignty-thinking-a-32095
-
OpenAI Limits GPT-5.6 Rollout at US Government’s Request
Enterprise Users Face Delayed Access as Trusted Partners Get Early Preview. OpenAI limited its release of GPT-5.6 to a short list of users after the Trump administration requested access to the model and the list of users. The staggered rollout came weeks after the government exerted emergency export controls over Anthropic’s Fable 5 model. First…
-
Weak Access Controls Leave Enterprise Networks at Risk
Barracuda researchers found that weak credentials and exposed remote services continue to fuel malware, botnet, and credential attacks. First seen on esecurityplanet.com Jump to article: www.esecurityplanet.com/threats/weak-access-controls-leave-enterprise-networks-at-risk/
-
Weak Access Controls Leave Enterprise Networks at Risk
Barracuda researchers found that weak credentials and exposed remote services continue to fuel malware, botnet, and credential attacks. First seen on esecurityplanet.com Jump to article: www.esecurityplanet.com/threats/weak-access-controls-leave-enterprise-networks-at-risk/
-
Cisco Adds NHI to Security Stack With Astrix, WideField Acquisitions
Cisco joins a growing list of security platform providers who are betting that securing the agentic workforce means turning identity into the primary control plane. First seen on darkreading.com Jump to article: www.darkreading.com/identity-access-management-security/cisco-adds-nhi-security-stack-with-astrix-widefield
-
Your First GRC Agent: A Red Teamer’s Walkthrough
AI won’t replace GRC analysts, but it can eliminate much of the repetitive work they do. Anecdotes walks through building an agent that continuously monitors controls, identifies evidence gaps, and opens remediation tasks. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/security/your-first-grc-agent-a-red-teamers-walkthrough/
-
New Linux pedit COW Exploit Enables Root Access by Poisoning Cached Binaries
A flaw in the Linux kernel’s traffic-control subsystem can let a local unprivileged user gain root on affected systems.CVE-2026-46331, nicknamed “pedit COW,” is an out-of-bounds write in the packet-editing action (act_pedit) that corrupts shared page-cache memory. A public, working exploit appeared within a day of the CVE assignment on June 16. Red Hat rates the…
-
Chinese-Speaking Hackers Deploy TinyRCT Backdoor Against Critical Energy Infrastructure
A Chinese-speaking threat cluster tracked as CL-STA-1062 has deployed a newly discovered .NET backdoor, TinyRCT, in targeted campaigns against government and critical energy infrastructure across Southeast Asia during 2025. The recent campaign combines common open-source tooling with bespoke malware. Operators consistently leverage publicly available utilities SoftEther VPN for tunneling, VNT and yuze for covert command-and-control,…
-
Hackers Use Malicious Minecraft Fabric Mods to Deploy LoaderClient and WeedHack Stealer
Hackers are weaponizing malicious Minecraft Fabric mods to deliver LoaderClient. This stage-one malware loader steals session data and hands it off to the WeedHack stealer through a fileless, blockchain-backed execution chain. The campaign stands out for its use of EtherHiding, where the command-and-control URL is pulled from an Ethereum smart contract instead of a conventional…
-
Modelplane: Open-source control plane for AI inference
Organizations that run open-weight models on hardware they own operate GPU fleets spread across clouds, neoclouds, and on-premise data centers. Each fleet handles model … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/06/26/modelplane-open-source-control-plane-ai-inference/
-
Stay ahead in the SOC: Contain threats with confidence and control
First seen on scworld.com Jump to article: www.scworld.com/resource/stay-ahead-in-the-soc-contain-threats-with-confidence-and-control
-
Aryon Secures $29M to Thwart Cloud Risks Before Deployment
Series A Funds Back Enforcement Controls That Block Insecure Resources Instantly. Aryon Security raised $29 million in Series A funding to help enterprises enforce security policies at cloud deployment, preventing misconfigurations, excessive permissions and insecure resources from reaching production environments across AWS, Azure and Google Cloud. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/aryon-secures-29m-to-thwart-cloud-risks-before-deployment-a-32069
-
The OT Segmentation Imperative: Why It Can’t Wait Any Longer
AI-Powered Attacks Make OT Network Segmentation a Business-Critical Control Industrial organizations are facing faster, more sophisticated attacks than ever before. As AI further accelerates cyberattacks, OT defenders can no longer rely on perimeter security alone. This blog explores why OT network segmentation is critical for limiting lateral movement. First seen on govinfosecurity.com Jump to article:…
-
Google releases new privacy controls for activity history, personalization
Google is rolling out new privacy controls for Search services and Google Play, giving you more control over saved history and personalized recommendations. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/google/google-releases-new-privacy-controls-for-activity-history-personalization/
-
LokiBot Malware Uses API Hashing and 3DES-Encrypted C2 to Hide Infostealer Activity
LokiBot, a long-lived infostealer first advertised in May 2015, continues to evolve. Recent samples demonstrate deliberate attempts to evade static detection and frustrate analysis by combining API hashing with 3DES-encrypted command-and-control (C2) configuration stored inside the binary. The result is a compact, stealthy loader that reconstructs and executes a traditional LokiBot payload while limiting observable…
-
DPRK-Linked macOS Implant Uses LaunchAgent Persistence and Python Stealer Module
The binary tracked as macOS.Gaslight as a Rust-based macOS implant and infostealer whose most novel features are analyst-directed prompt injection and a hardened Telegram-based command-and-control (C2) channel. We assess with high confidence that macOS.Gaslight aligns with DPRK-linked macOS activity clustered around BONZAI and AIRPIPE signatures. macOS.Gaslight is ad hoc signed, carries the identifier endpoint-macos-aarch64-5555494492fc075f441637fb9d894913dde3a2ea, and…
-
Anthropic Launches Claude Tag, Bringing AI Agents Into Slack
Anthropic launched Claude Tag in Slack, giving enterprise teams an AI agent with shared context, admin controls, logs, and spend limits. The post Anthropic Launches Claude Tag, Bringing AI Agents Into Slack appeared first on TechRepublic. First seen on techrepublic.com Jump to article: www.techrepublic.com/article/news-anthropic-claude-tag-ai-agent-slack/