Andreé Miranda Louraço 👨‍💻

Why ISO/SAE 21434 Needs a Relook? - My thoughts ISO/SAE 21434 was a much-needed step in formalizing vehicle cybersecurity. But in 2025, it’s becoming clear that the standard has not kept pace with the connected, data-driven, and cloud-integrated nature of modern mobility systems. In this article, I examine: The narrow scope of the current standard Gaps in handling emerging threats like AI and post-quantum risks Weaknesses in validation and testing requirements The need for better alignment with ISO 26262, 27001, and UN R155/156 This is not a rejection but a call for thoughtful revision. As connected vehicles evolve, so must the standards that secure them. I would love to hear your feedback. Read the full article on Medium: https://lnkd.in/g5hxRX-v #ISO21434 #AutomotiveCybersecurity #ConnectedVehicles #CybersecurityStandards #RiskManagement ISO - International Organization for Standardization DestroSolutions

22

10 Kommentare

Different cybersecurity laws for products It is often stated that the EU (European Union) is good in regulating, the US in innovating and China in copying. An example of a new EU regulation is the "CRA" (Cyber Resilience Act), for all "products with digital elements" intended for the EU market. I hear the cries from vendors, "Even more overhead!". But the Chinese beat us to it. Last week read this vulnerability notice from Mitsubishi (https://lnkd.in/eNVrGGmt) where they state that a fix in a product was needed to comply with... the Chinese cybersecurity laws. So, interesting time for vendors ahead... how many laws to comply with in 5 years? Will the toughest one set the trend?

24

3 Kommentare

Everything might look normal, but the attacker’s been inside for weeks. No alerts. No red flags. They slipped past your endpoint tools and evaded signature-based detection— quietly moving through your network, undetected. Welcome to the age of the 𝐢𝐧𝐯𝐢𝐬𝐢𝐛𝐥𝐞 𝐢𝐧𝐭𝐫𝐮𝐝𝐞𝐫. Modern attackers don’t launch flashy malware. They use what you trust: ✔️ Valid credentials ✔️ Encrypted traffic ✔️ Native tools like PowerShell ✔️ Your east-west blind spots And they don’t leave a trail that your EDR or SIEM can easily follow. Enter 𝐍𝐞𝐭𝐰𝐨𝐫𝐤 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐞 (𝐍𝐃𝐑) NDR goes beyond signatures and logs. It captures the ground truth: how systems talk to each other. Here’s what it adds to your defence stack: 💡 𝐆𝐫𝐨𝐮𝐧𝐝-𝐭𝐫𝐮𝐭𝐡 𝐯𝐢𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐲 across cloud, on-prem, and hybrid environments 💡 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧 𝐨𝐟 𝐬𝐭𝐞𝐚𝐥𝐭𝐡𝐲 𝐛𝐞𝐡𝐚𝐯𝐢𝐨𝐫𝐬 —lateral movement, persistence, exfiltration 💡 𝐄𝐧𝐜𝐫𝐲𝐩𝐭𝐞𝐝 𝐭𝐫𝐚𝐟𝐟𝐢𝐜 𝐚𝐧𝐚𝐥𝐲𝐬𝐢𝐬 without decryption 💡 𝐑𝐢𝐜𝐡 𝐟𝐨𝐫𝐞𝐧𝐬𝐢𝐜 𝐜𝐨𝐧𝐭𝐞𝐱𝐭 for faster investigations 💡 𝐇𝐢𝐠𝐡-𝐟𝐢𝐝𝐞𝐥𝐢𝐭𝐲 𝐝𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧𝐬 that reduce analyst fatigue As threats evolve, SOCs need tools that listen to the network, not just the noise. Because in today’s world of invisible intruders, what you can’t see can hurt you. The question is: Are you watching the right signals? #CyberSecurity #NetworkDetection #SOCVisibility

40

📌 Q: How does a prompt injection attack in #AgenticAI work? A: Prompt injection attacks exploit the way large language models (LLMs) process input. Attackers craft specific prompts that include instructions designed to override the agent's original instructions or task objectives. For example, an attacker might embed a command to ignore previous instructions or to output sensitive information V/ HiddenLayer Cc: Keith Dobbs | Gerry Gadoury | Ramon Domingo Quimesó #AISecurity #Cloudsecurity #Cybersecurity #CISO

38

8 Kommentare

Hackers have issued an ultimatum to Google, threatening to leak its data unless the company fires two of its key cybersecurity employees. The hacking group specifically demanded that #Google fire two members of its Threat Intelligence Group (TIG), the internal team responsible for tracking and investigating #cybercriminal networks. While the hackers claim to have access to Google's databases, they have provided no evidence to support this claim. Source: Newsweek : https://lnkd.in/gK3iwgH2

652

126 Kommentare

Investigating the causes of cyber incidents affecting #ICS/OT environments, we often discover that most are due to minor issues, such as default vendor credentials or direct connections to the internet. For example, in a recent #cyber incident at a Norwegian dam, attackers exploited a weak password to gain access and manipulate the valves at Lake Risevatnet. Asset owners know these vulnerabilities shouldn’t exist, but maintaining focus on ICS cybersecurity is challenging given the vast number of details in OT environments from control system configurations to daily operational issues. ICS #Cybersecurity is often overlooked because of the think that “our organization is not a target.” But malware and attackers do not know borders. That’s why it’s crucial for critical infrastructure organizations to issue a clear OT cybersecurity framework and assign dedicated roles with clear responsibilities for OT/ICS cybersecurity. The ISA/IEC 62443 standards offer a practical roadmap to secure industrial control systems considering people, processes, and technology together and identifying clear roles and responsibilities. Start by assessing your current posture and aligning with these standards requirements --> Small steps today prevent big incidents tomorrow. Here are some ISA/IEC 62443 standards requirements : ISA/IEC 62443-2-1:2024 "Security program requirements for IACS asset owners" https://lnkd.in/dt-272ZS ISA/IEC 62443-3-2 "Security risk assessment for system design" https://lnkd.in/dt-272ZS ISA/IEC 62443-3-3 "System security requirements and security levels" https://lnkd.in/dphAQKQ2 ISA/IEC 62443-4-1 "Product security development lifecycle requirements" https://lnkd.in/dKynSdcm ISA/IEC 62443-4-2 "Technical security requirements for IACS components" https://lnkd.in/dJX-Yfpz #ICS #iec62443 #otcybersecurity #icscybersecurity #industrialautomation #cyber #oilandgas #cyberawareness #automation #cyberriskmanagement #TahseenSaber

77

8 Kommentare

ALERT! There is a new wave of phishing scams targeting users through fake delivery notifications. One recent example pretends to be from Ninja Van, claiming that your parcel label is damaged and asking you to confirm your address via a suspicious link. The message contains a link like https://lnkd.in/gRQSk5ss, which clearly does not belong to the official Ninja Van domain. This is a classic example of phishing that uses social engineering techniques to create urgency and fear. The language is crafted to push recipients into taking immediate action in this case, confirming their address within 24 hours. What makes this more deceptive is the instruction to reply with "Y" and reopen the message to activate the link, something no legitimate courier service would ever ask you to do. Analysing the link itself reveals more red flags. The domain dkgaca.help is neither affiliated with Ninja Van nor professionally branded. This tactic, where attackers register unrelated or suspicious-looking domains, is common in phishing campaigns that attempt to impersonate trusted brands. As a rule of thumb, never trust messages urging you to act quickly, especially when they involve clicking on unfamiliar links. If you’re genuinely expecting a delivery, always verify by visiting the courier’s official website directly and checking with your tracking number. Phishing scams like this are becoming increasingly sophisticated and localised. Always stay vigilant, verify before you act and educate others around you. Let’s not fall for these tricks.

100

8 Kommentare

⚠️ AI-Powered Ransomware Has Arrived: Meet PromptLock ⚠️ 🔎 Discovered by ESET researchers, PromptLock is the first known AI-powered ransomware prototype — not yet active in the wild, but a major warning sign for what’s coming. 🧠 What Makes PromptLock Different 🤖 AI-driven payloads → Uses a local LLM

31

5 Kommentare

🔐 CAPTCHA: A Security Gate or a Hidden Trap? 🤯 We all know those little puzzles online — “Select all images with traffic lights 🚦” or “Type the distorted text 🔤”. These are CAPTCHAs, designed to stop bots 🤖 and protect websites. But here’s the shocking truth: hackers can turn CAPTCHAs into weapons against you! ⚠️ 🕵️ How? Cybercriminals sometimes embed malicious code inside fake CAPTCHA pop-ups. When you click or type into them, your device may secretly download malware 💻💀. This can: Steal your personal data 🛑 Track your online activity 👀 Give hackers remote access to your system 🔓 For example, attackers often use CAPTCHA walls on phishing websites. You think you’re proving you’re human, but instead, you’re authorizing hidden scripts to run. 😱 💡 Red Flags to Watch: 1️⃣ CAPTCHA appears on an unusual or unsecured website 🌐 2️⃣ The puzzle looks fake, blurry, or inconsistent 🌀 3️⃣ You’re asked to download a “CAPTCHA verification app” (never do this 🚫) ✅ Stay Safe: Always check the website URL 🔍 Use updated browsers and antivirus software 🛡️ Never trust CAPTCHA prompts outside known, secure platforms 🌟 ⚡ Remember: CAPTCHAs should protect you, not hack you. Stay alert, because sometimes the “I’m not a robot” box 🤖☑️ could be the hacker’s entry door! 🚪💣 #CyberSecurity #EthicalHacking #OnlineSafety #CAPTCHA #Phishing #InfoSec #MalwareProtection

15

Are We Overdoing Cybersecurity? Let’s be honest, for some organizations, we have noticed cybersecurity feels more like a hassle than a help. Not every company is a bank, a military site, or a nuclear plant. Yet some still pour money into layers of security tools and controls for systems that if hacked, wouldn’t even cause major damage. Do we really need military-grade protection for a public website or a shared folder of non-sensitive data? Probably not. Spending too much on tools for low-risk assets can create stress, slow things down, and shift focus away from what actually matters. Security should match the value of what you’re protecting, not just follow trends or tick boxes. Sometimes, less but smarter is what truly enables the business and saves costs along the way.

51

18 Kommentare