Stronger AI Safety Requires Peeking Inside the ‘Black Box’
Researchers propose focusing on identification of certain cognitive elements in LLMs that indicate when AI systems may take an unwanted action.
-
-
Claude AI Just Cracked a Post-Quantum Test Scheme and Found a Faster 7-Round AES Attack
Anthropic says Claude Mythos Preview helped derive an end-to-end key-recovery attack against HAWK-256 and a 200- to 800-fold speedup for an attack on seven-round AES-128. The HAWK attack exploits a previously unused symmetry in the lattice behind the signature scheme. Anthropic’s released implementation gives an expected end-to-end runtime of about three hours and 42 minutes…
-
‘Certighost’ Flaw Haunts Microsoft Active Directory Certificates
Microsoft patched a high-severity vulnerability earlier this month that allows a threat actor to escalate privileges and compromise an AD environment.
-
Tengu Botnet Reboots Compromised Linux Devices When Defenders Kill Its Process
A new Mirai-derived botnet called Tengu can use a compromised Linux device’s hardware watchdog to trigger a reboot when defenders kill its main process. If that happens, Tengu’s other persistence mechanisms get another chance to relaunch it. Nozomi Networks Labs observed the dropper reaching its honeypots through Telnet credential brute force. Tengu supports 25 distributed…
-
24,650 Internet-Exposed BMCs Disclose IPMI Password Hashes Before Login
Cybersecurity researchers have sounded an alert after finding more than 36,000 Baseboard Management Controller (BMC) management interfaces exposing Intelligent Platform Management Interface (IPMI) protocol to the public internet. Of the 36,872 internet-exposed server-management interfaces running IPMI, 24,650 have been found to disclose password-derived authentication hashes before login due to
-
JFrog Confirms OpenAI Models Exploited Artifactory Zero-Day Before Hugging Face Breach
JFrog has confirmed that OpenAI models exploited a zero-day in self-hosted Artifactory while trying to reach the open internet from a sealed evaluation environment. Artifactory is JFrog’s software repository manager. OpenAI says the models then escalated privileges and moved laterally until they reached an internet-connected node. JFrog says it has since developed and released fixes…
-
Critical OpenWrt DHCPv6 Flaw Could Let Unauthenticated Attackers Run Code as Root
OpenWrt has shipped version 24.10.8 to close a critical DHCPv6 stack overflow and a wider set of remotely triggerable flaws in network services enabled by default. The critical issue, tracked as CVE-2026-53921 and rated 9.8 on CVSS 3.1 in OpenWrt’s GitHub advisory, lets an unauthenticated attacker able to reach the DHCPv6 server overwrite a stack…
-
Former Citigroup CISO Blauner on What Makes A Great Security Leader
The cybersecurity pioneer discusses the evolution of the CISO role, AI’s impact on careers, and why operational resilience is the profession’s next frontier.
-
Nimbus Manticore Deploys NightLedger and Turns Victim Systems Into Covert Relays
The Iranian state-backed hacking group tracked as Nimbus Manticore (aka GalaxyGato, Mirage Kitten, Smoke Sandstorm, Subtle Snail, and UNC1549) has been attributed to a fresh set of attacks targeting entities across the Middle East, Africa, and South Asia. The intrusions involve the use of a previously undocumented Windows backdoor called NightLedger and two custom WebSocket…
-
Critical TeamCity Flaw Could Let Attackers Run OS Commands Without Logging In
JetBrains is urging customers of on-premise versions of TeamCity to update to the latest version following the discovery of a critical security issue that could result in arbitrary code execution. The vulnerability, assigned CVE-2026-63077 (CVSS score: 9.8), affects all TeamCity On-Premises versions. It has been addressed in versions 2025.11.7 and 2026.1.3. TeamCity Cloud instances have…