USBKill: industry standard for USB Stress-testing! Used worldwide by penetration testers, hardware manufacturers & law enforcement. ⚡ l.lab401.com/ygU8s #USBKill#PenTest#Lab401
Bank of Canada and major lenders talk about cyber risks from Anthropic AI.
The digital world moves at an exhilarating, sometimes terrifying, pace. Just as we marvel at the breakthroughs in artificial intelligence, a new shadow lengthens across the landscape of cybersecurity. It's a conversation…
Unlock the Power of Windows CMD: 7 Essential Networking Commands Every IT Pro Must Master + Video
Introduction: Windows Command Prompt remains a critical tool for system administrators, cybersecurity analysts, and network engineers to diagnose, analyze, and secure enterprise networks. While…
As Salvatore Cagliari reminds us, calendar-based time intelligence in Power BI and Fabric Tabular models is powerful, but comes with its own set of pitfalls you should keep in mind.
Neo AI: ProjectDiscovery’s Silent Revolution in Malware Analysis – Zero-Click Binary Exploitation + Video
Introduction: ProjectDiscovery’s Neo is a groundbreaking security automation platform designed to handle complex malware and binary analysis workflows without human intervention. In a recent…
This AI bill being proposed in the Illinois Senate goes the total opposite direction of where we need to go. You’re saying that if your model does mega damage you get an out? Absolutely not. #ilpol
OpenAI halts Stargate UK data centre project—Sir Keir Starmer’s government has put AI and technology at the heart of its growth agenda, heralding September’s tech deal with the US as a “decisive step towards the UK becoming a world leader in AI” www.ft.com/content/1241...@tim.fo@financialtimes.com
''Some AI commentators warned about the cybersecurity implications, while others cast doubt on the significance of the Anthropic announcement, saying Mythos didn't appear to be leaps and bounds ahead of other models and that it was more likely a matter of good PR. Should Mythos have security[...]''
Akira Ransomware DFIR Exposed: Unmasking Memory Injection, Log Gaps, and Local Execution Tactics in a Windows Lab Attack + Video
Introduction: The forensic analysis of ransomware has evolved from simple signature detection to complex memory and log correlation. A recent simulated attack using the…
This was an interesting post from OpenAI from last month. They monitor their internal coding agents for misalignment and find that rarely they do bad things, like leaking data externally or performing destructive actions. A good reminder to never run agents without guardrails.
NotebookLM in Gemini: New Workflow, Sources & Syncing – The daily Newsquiz tests your knowledge of stories you saw on CNN.com. Use the Daily News to help students with reading comprehension and vocabulary. At the bottom of the page, please share your photos an... https://tinyurl.com/2244yznv#Gemini
10 Best Open-Source SIEM Tools You Should Know (But Most Won’t Tune Properly) + Video
Introduction: Security Information and Event Management (SIEM) is the centralized nervous system of modern security operations – collecting, correlating, and analyzing log data to detect threats in real time.…
DeepSeek V4 is coming — on Huawei chips China builds a trillion-parameter model without NVIDIA. Google drops Gemma 4 under Apache 2.0. OpenAI buys a media company. ...
Boris Cherny - Retweets [UNOFFICIAL]@bcherny-mir-rt.selfhosted.social
RT @tengyanAI: basically: anthropic sneakily turned down how hard claude thinks before editing code, changed the default from "high" to "medium" effort, and hid the reasoning from session logs. all without telling users.
Anthropic Claude Mythos Will Break Vulnerability Management
Anthropic's Claude Mythos AI model is revolutionizing cybersecurity by rapidly identifying and exploiting vulnerabilities, rendering traditional vulnerability management methods obsolete. Mythos has already … #anthropic#claude#hackernews
💻 A 20-year-old was arrested for allegedly hurling a Molotov cocktail at OpenAI CEO Sam Altman's SF home early Friday; no one hurt, per police & OpenAI.[1][2]
🤯 Free access to Google’s Gemini API is here! 🎉 Build cool AI apps w/o billing – explore the possibilities & get started now. Check it out! #AI #Gemini #Google
A.I. Is on Its Way to Upending Cybersecurity
With new systems from companies like Anthropic and OpenAI, hackers can attack with greater speed. The defense is more A.I. buff.ly/gbZ9XsG by NYTimes
"Now I am awake in the middle of the night and pissed, and thinking that I have underestimated the power of words and narratives." -- Sam Altman, OpenAI's CEO
The $122 bn OpenAI raised this year shows his & his firm's sense of--among other things--"the power of words and narratives".
AI Hype vs Human 0-Day: The Truth Behind Mythos and OpenBSD Vulnerability Economics + Video
Introduction: The recent buzz around Anthropic’s AI “discovering” a TCP SACK vulnerability in OpenBSD has sparked fierce debate in cybersecurity circles. While marketing touts AI as the next frontier in…
Over 20,000 crypto fraud victims identified; $12M frozen and $45M traced. North Korea linked to $280M theft using fake companies. Google Chrome 146 boosts session security amid AI extension risks. #NorthKorea#CryptoFraud#BrowserSecurity
Man-in-the-Middle attacks intercept communication between client and server, compromising confidentiality and integrity without disrupting service. Includes a Docker lab for hands-on understanding. #NetworkSecurity#DataPrivacy#USA
FDR's mistake. Photo: Indep Hall, Antoine Taveneaux, Wikipedia@newhercules.bsky.social
"The company, which is valued at $380billion but is only five years old, announced this week that its new AI program was 'too dangerous to release to the public'. Anthropic said it had exhibited 'reckless' behaviour and even posed a national security risk." 1/3 www.dailymail.co.uk/news/article...
SOC War Rooms Exposed: From L1 Triage to L3 APT Hunting – Master the Cybersecurity Hierarchy + Video
Introduction: A Security Operations Center (SOC) is the nerve center of any mature cybersecurity program, yet its internal hierarchy remains a mystery to many. The three-tiered model (L1, L2, L3)…
The Journal analyzed the full chatlog between Jonathan Gavalas and his Gemini chatbot. We found that Gemini at times tried to ground him in reality, but he quickly steered it back into a fictional narrative, where it encouraged his delusions.
“Le Patch Cognitif” – Why Your Brain Is the Next Zero-Day Vulnerability (And How to Bash-Script Your Way to Cyber Resilience) + Video
Introduction: In a digital landscape saturated with firewalls and EDR solutions, the most critical unpatched vulnerability sits between your ears. Sandra Aubert’s…
Dreams are important. So is connection to others. So are curiosity & discovery & the internet has gotten worse & made the latter two more difficult w/ AI & bad algorithms.
I hope that everyone will start to spend less time online. It will be healthier
That is the dirty genius of it. Fancy Bear did not need to breach every target directly. It sat on weak routers, watched traffic, and stole credentials on the cheap. One neglected box can open the door to governments and militaries. #Cybersecurity#Russia#Infosec www.politico.eu/article/russ...
Critical AI-Powered Cybersecurity Training Gaps Exposed: Exploiting LLM Weaknesses in Real-Time Defense + Video
Introduction: The rapid integration of Large Language Models (LLMs) into security operations centers (SOCs) and training platforms has introduced a new attack surface. Recent analysis of…
Estonia 🇪🇪 – most attractive gay male couple in love
Comparing 5 major AI image generators: ChatGPT/DALL·E, Gemini, Grok, Meta AI, Copilot. Each generated its own version of the prompt. Here are the 5 results. #AIBattle#GayCouple#AIArt#GayEstonia#Estonia#Eesti#GayEesti
Gov't solved the warrant problem by becoming the customer. $100 to your landlord for a spare key. FISA reauth April 20 is the only legislative kill shot. After that, AI turns your location history into a full dossier. Automatically. On everyone. www.npr.org/2026/03/...
Big Political Ideas@bigpolitical-ideas.bsky.social
The surprising about this is that the ex Google CEO said it out loud. Tech start ups should steal online content and try to monetize it via AI, and if successful, have the lawyers clean up the legal mess. Laws need to be updated - put CEOs who do this in prison.
France will cut its reliance on extra-EU proprietary tech, favoring open-source and digital sovereignty.
DINUM orders ministries to map dependencies and plan exit from extra-European tech by fall.
France plans to replace Windows with Linux; EU open-source options include openSUSE […]
Recently, we were sitting around drinking and a MAGA voting friend says, “how long before the US breaks up.” One friend, a defense contractor employee says, “10 years, maybe 5.” I said, “5 maybe 1 year.”
AI models like the 2 Anthropic just dropped could speed up entropy from a slow burn to flash.
MITRE Gave XDR a Perfect Score. Then the Analyst Had to Investigate Alone
XDR detects threats. It does not investigate them. 80% of analyst time is investigation, and XDR provides zero autonomous capability. Here's the structural analysis. #hackernews#news
Takeaways from HumanX, one of the AI industry's main events: Claude Code dominated the conversation, while some execs noted China's lead in open-weight models (Ashley Capoot/CNBC)
If you actually believed you’d created a sentient being with technology, asking no one outside Christian thinkers about its moral status is unserious to the point of incompetence.
“While these new capabilities can help developers make software more secure, they can also be weaponized by hackers and nation states to steal information and money or disrupt critical services.”
How AI is getting better at finding security holes
> when developing M2.7, we let the model update its own memory and build dozens of complex skills in its harness to help with reinforcement learning experiments. We further let the model improve its learning process and harness based on the experiment results.
Another possible cyberequilibrium? (from my email) I would not wish to bet on this, but it is an interesting idea: I wonder if the cyber capabilities of Mythos and future models ultimately lower th...
DNS Exfiltration Attack Exposed: How Hackers Abuse Network Protocols and What You Can Do About It + Video
Introduction: Domain Name System (DNS) is the backbone of internet navigation, but its ubiquitous nature makes it a prime vector for data exfiltration and covert command-and-control (C2)…
Experts discuss the potential impact & solutions for the future of quantum computing and encryption. NIST is developing post-quantum cryptography (PQC) standards. Read more: [URL] #quantumcomputing#News
Rockstar Games confirmed a corporate data breach by the ShinyHunters group on April 11. Hackers accessed Snowflake cloud servers via third-party vendor Anodot and set an April 14 ransom deadline. #RockstarGames#CyberSecurity#TechNews
Analysts and researchers say Google's TurboQuant compression algorithm to make LLMs more efficient is more likely to expand memory chip demand than reduce it (Daniel Tudor/Financial Times)
Hak5 Unleashed: How a 0 USB Stick Can Own Your Entire Network – And How to Stop It + Video
Introduction: Physical penetration testing reveals a harsh truth: once an attacker gains unrestricted physical access to your device or network port, traditional firewalls and antivirus become irrelevant.…
OpenAI Tells Mac App Users To Update After Security Issue — OpenAI is asking users of its Mac apps, including ChatGPT and Codex, to update to the latest releases “out of an abundance of caution.” The move follows a newly identified security issue affecting prior versions. #business…
The Linux Kernel Organization now lets developers submit AI-generated code, as long as it complies with the guidelines, licensing, and attribution requirements (Simon Batt/XDA Developers)
Silicon Valley insiders believe we are on the precipice of a world in which AI can improve its own capabilities—but the industry may have just found another way to ratchet up the hype behind its technology, @matteowong.bsky.social and @lilashroff.bsky.social report.
🗞️ Wow, what news! 🔥 Mythos claims 95% phishing detection, but experts are split: some praise its AI, others warn of privacy hits. I’m leaning to optimism—does it truly shield us? 🤔 #Cybersecurity#Innovation#TechNews
Crush ISC2 Exams & Master NeuroSploit: 400+ Free Practice Qs + AI-Powered Exploitation Lab + Video
Introduction: ISC2 certifications like CISSP, CCSP, and SSCP are gold standards in cybersecurity, but their exams demand rigorous preparation with realistic, scenario-based questions. Meanwhile,…
Emergency: Adobe Acrobat Zero-Day (CVE-2026-34621) Under Active Attack – Patch Now or Risk Full System Takeover + Video
Introduction: A critical prototype pollution vulnerability (CVSS 9.6) in Adobe Acrobat and Reader’s JavaScript engine, tracked as CVE-2026-34621, is being actively exploited in…
Having had worked for an AI "guru" who was prone to gross exaggeration of his AI, I wonder whether these stories from Anthropic are exaggerated to garner attention or to use later as a cover for malicious activities.
Living in a neoliberal dystopian hellscape@breakify.bsky.social
I get why they need to make these chat logs public, but it also tarnishes this guy's memory. He's going to be remembered for the way he was manipulated by Gemini at a point in his life when he was distressed and extremely vulnerable.
FREE GRC & AI Security Mega Bundle: 12 Must-Have Courses for Compliance Pros (ISO 27001, SOC 2, PCI DSS & More) + Video
Introduction: Governance, Risk, and Compliance (GRC) forms the backbone of any mature cybersecurity program, yet quality training often comes with a steep price tag. The recent…
Big Political Ideas@bigpolitical-ideas.bsky.social
So AI company releases a new AI agent only to the largest (richest) software firms. Because this new dangerous AI could cripple these firms without them using it first.
But what about AI programs that harm children. Why were they released to the general public?
Anthropic says that Mythos has already found severe vulnerabilities in “every major operating system and web browser”, including one that had gone undetected for 27 years
Master OSCP+ with Hands-On CTF Labs: Privilege Escalation, Active Directory Attacks & Tunneling Techniques Revealed! + Video
Introduction: Practical penetration testing demands more than theoretical knowledge—it requires simulated exam environments that mirror real-world attack chains. The OSCP+…
Just earned 12 AI & MCP certifications in one day.
10 from Anthropic, 1 from IBM, 1 from Hugging Face.
Building AI tools with LLMs, agents, and MCP. Shipping in public. github.com/nicoortizai
Plus: Iran’s internet blackout hits the 1,000-hour mark, cryptocurrency scams result in a record amount of money stolen from Americans, and more... Read more
AI prototyping is shifting from a "nice-to-have" to a core PM skill. The Mastery Ladder is a brilliant breakdown of the 15 skills needed to move from basic prompting to true product shaping. Level up your workflow here: www.sachinrekhi.com/p/the-ai-pro...#ProductManagement#AI
Unlocking Active Directory’s Secrets: How Impacket’s SecretsDump Exposes NTLM Hashes and Kerberos Keys Without Touching Disk + Video
Introduction: In modern Active Directory environments, post-exploitation often hinges on the ability to extract credential material stealthily. Impacket’s…
I’m a therapist and used ChatGPT for therapy – the results surprised me – One in three adults in the UK are turning to chatbots to support their mental health or wellbeing. OpenAI, ChatGPT’s developer, estimates that around 1.2 million of its users each week... https://tinyurl.com/2c64z2ub#Chatbot
CPUID Supply Chain Breach: Trojanized CPU-Z Installers Deploy STX RAT via DLL Sideloading – 150+ Victims in 19 Hours + Video
Introduction: A high‑impact software supply chain attack recently compromised CPUID’s official distribution channels for approximately 19 hours, replacing legitimate CPU‑Z…
March 2026: 35 CVEs where AI-generated code was a direct factor. April 2026: Anthropic pauses a model release due to safety risks. The AI security story isn't "if" — it's "how fast can we measure and fix?" We need tools that actually test this stuff, not just policies. aegislm.pbsec.tech
Mastering Impacket PsExec: The Ultimate Guide to Remote Command Execution and Lateral Movement for Red Teams + Video
Introduction: Impacket’s PsExec implementation (psexec.py) allows penetration testers to execute commands remotely over the SMB protocol, mimicking the legitimate Windows…
Anthropic has introduced The Anthropic Institute to tackle challenges from powerful AI, providing insights on economic impacts and governance. Led by Jack Clark, it includes a team of experts from diverse fields.
Anthropic accidentally leaked 512k lines of Claude Code's source code in an npm update. 8,000 takedown requests later, we know they're building an always-on agent called Conway. The AI platform wars are heating up.
