A critical security vulnerability in LangChain, one of the world’s most widely deployed AI frameworks, enables attackers to extract environment variable secrets and, through a serialization injection flaw, potentially achieve code execution. The vulnerability, identified as CVE-2025-68664, affects the core…
Category: EN
Parrot 7.0 Released with New Penetration Testing and AI Tools
Parrot OS 7.0, codenamed Echo, launches as a complete system rewrite based on Debian 13, bringing KDE Plasma 6, Wayland by default, and fresh penetration testing tools, including a dedicated AI category. This release emphasizes lightweight theming and community-driven spins,…
From AI to cyber risk, why IT leaders are anxious heading into 2026
Cybersecurity threats are shaping IT planning for 2026, with AI maturity and regulation emerging as another major source of disruption, according to a global survey from Veeam. Veeam surveyed 250 senior IT and business decision-makers worldwide to understand how they…
The next big IT security battle is all about privileged access
Leostream predicts changes in Identity and Access Management (IAM) and Privileged Access Management (PAM) in 2026 driven by new realities of cybersecurity, hybridization, AI, and more. Passwordless moves from pilot to production In 2026, passwordless authentication will shift from isolated…
The Ransomware Ecosystem: An Encore Holiday Episode
Jim takes a break for some R&R during the holidays and shares his favorite podcast episodes from the year. He acknowledges that some listeners might have heard these episodes already, while others may find them new. The podcast’s production is…
Utair – 401,400 breached accounts
In August 2020, news broke of a data breach of Russian airline Utair that dated back to the previous year. The breach contained over 400k unique email addresses along with extensive personal information including names, physical addresses, dates of birth,…
Critical Langchain Vulnerability Let attackers Exfiltrate Sensitive Secrets from AI systems
A critical vulnerability in LangChain’s core library (CVE-2025-68664) allows attackers to exfiltrate sensitive environment variables and potentially execute code through deserialization flaws. Discovered by a Cyata researcher and patched just before Christmas 2025, the issue affects one of the most…
Are IT managers truly satisfied with current AI security measures
Why Are Non-Human Identities Crucial in AI Security? Have you ever considered how closely your organization’s cybersecurity posture is tied to its management of Non-Human Identities (NHIs)? Where digital threats increasingly target machine identities, the role of NHIs in AI…
What makes a business free to choose the best AI security
How Can Organizations Effectively Manage Non-Human Identities (NHIs) to Enhance Cloud Security? Have you ever wondered how machine identities fit into the broader framework of cybersecurity? Managing Non-Human Identities (NHIs) is pivotal for enterprises aiming to bolster their cloud security…
How do secrets management tools deliver value to organizations
What Are Non-Human Identities and Why Do They Matter in Cybersecurity? Security teams often grapple with a pivotal question: How can organizations effectively safeguard Non-Human Identities (NHIs)? NHIs play a crucial role in cybersecurity by acting as unique identifiers for…
Can Agentic AI operate independently within secure parameters
How is Agentic AI Revolutionizing the Management of Non-Human Identities? Have you ever considered the autonomous potential of Agentic AI in managing Non-Human Identities (NHIs) within your organization? With industries increasingly rely on machine identities for efficient operations, securing these…
What Makes a Successful GRC Team? Roles, Skills, & Structure
Key Takeaways A GRC team is responsible for defining how requirements are interpreted, how risks are assessed and tracked, and how accountability is maintained across the organization. While the GRC team provides central oversight, effective execution depends on coordination with…
Five-year-old Fortinet FortiOS SSL VPN vulnerability actively exploited
Fortinet reported active exploitation of a five-year-old FortiOS SSL VPN flaw, abused in the wild under specific configurations. Fortinet researchers observed “recent abuse” of a five-year-old security vulnerability, tracked as CVE-2020-12812 (CVSS score: 5.2), in FortiOS SSL VPN. The vulnerability is…
Google Now Allows Users to Change Their @gmail.com Email Address
For years, one of the most persistent frustrations for Google users has been the inability to alter their primary email address without creating an entirely new account. Whether you are stuck with an unprofessional handle created in high school or…
Unpatched FortiGate Security Flaw Allows Attackers to Bypass 2FA Controls
A critical authentication bypass vulnerability in FortiGate devices enables threat actors to circumvent two-factor authentication (2FA) protections through case-sensitive username manipulation. The flaw, tracked as CVE-2020-12812, affects organizations with specific LDAP integration configurations and remains exploitable on unpatched systems. The…
Shinhan Card Faces Regulatory Review Over Internal Data Sharing Incident
Shinhan Card, one of South Korea’s largest credit card companies, has disclosed a data leak involving the personal information of approximately 192,000 merchants. The company confirmed the incident on Tuesday and said it has notified the Personal Information Protection…
US Justice Department Seizes Web Domain Linked to Large-Scale Bank Account Takeover Fraud
The U.S. Justice Department (DoJ) on Monday revealed that it has taken control of a web domain and its associated database that were allegedly used to support a criminal operation aimed at defrauding Americans through bank account takeover fraud.…
Best of 2025: New Akira Ransomware Decryptor Leans on Nvidia GPU Power
A software programmer developed a way to use brute force to break the encryption of the notorious Akira ransomware using GPU compute power and enabling some victims of the Linux-focused variant of the malware to regain their encrypted data without…
Cookies Explained: Accept or Reject for Online Privacy
Online cookies sit at the centre of a trade-off between convenience and privacy, and those “accept all” or “reject all” pop-ups are how websites ask for your permission to track and personalise your experience.Understanding what each option means helps…
ThreatsDay Bulletin: Stealth Loaders, AI Chatbot Flaws AI Exploits, Docker Hack, and 15 More Stories
It’s getting harder to tell where normal tech ends and malicious intent begins. Attackers are no longer just breaking in — they’re blending in, hijacking everyday tools, trusted apps, and even AI assistants. What used to feel like clear-cut “hacker…