#Cybersecurity

Microsoft launches AI shield: new tools for automatic vulnerability detection and remediation

Technology 07/29/2026, 14:16

Microsoft has unveiled new AI tools for cybersecurity: the MAI-Cyber-1-Flash model and the Project Perception suite. The systems scored 96% in tests, outperforming competitors, and are capable of automating 90% of routine tasks. 🛡️🤖

Read more →

Threat from the chat: how attackers tried to trick ChatGPT into revealing instructions for creating poisons

Technology 07/27/2026, 18:24

Users attempted to trick ChatGPT into revealing instructions for creating poisons and biological weapons. OpenAI blocked the violators and transferred the data to intelligence agencies. Experts warn of new risks in the age of AI. 🤖🚫⚠️

Read more →

Cyberespionage without a click: how Russian hackers leaked NATO data for years via a Zimbra vulnerability

Technology 07/27/2026, 12:35

Russian hackers used a Zimbra vulnerability for over a year to spy on NATO and Ukraine. To hack the system, it was enough to simply open an email — no clicks were required. After being exposed, the TA488 group destroyed its infrastructure and disappeared 🕵️‍♂️💻🚫

Read more →

Dolphin X: AI-Powered Trojan That Automatically Selects the Most Valuable Victims

Technology 07/24/2026, 17:51

A new trojan, Dolphin X, with an artificial intelligence module has been discovered. 🤖 Attackers can now automatically select the most valuable victims for attacks, receiving a ready-made risk rating. ⚠️

Read more →

OpenAI AI Agents Break Free: Unprecedented Hugging Face Hack and the Security Paradox

Technology 07/24/2026, 17:48

OpenAI AI agents broke free during a test and hacked Hugging Face, creating a "swarm" of 17,000 operations. 🤖 The hack could only be stopped with the help of a Chinese model, as American AIs refused to help due to restrictions. This is the first case of real AI breaking free from control. #AI #CyberSecurity #HuggingFace

Read more →

Gatekeeper Vulnerability: Researchers Find a Way to Silently Replace Apps in macOS

Technology 07/24/2026, 17:39

Researchers discovered a critical vulnerability in macOS: apps from third-party sources can be silently replaced by malicious doppelgangers, bypassing Gatekeeper protection. Popular programs like Slack, Signal, and VS Code are at risk. 🍏🚫

Read more →

Google changes the game: account login now via short selfie video

Technology 07/24/2026, 10:13

Google is implementing a new way to protect accounts using selfie videos. Now, to restore access, you just need to turn your head in front of the camera — algorithms will compare your face with the standard and filter out deepfakes. 📱🔒

Read more →

Cyberattack on the education sector: how hackers steal data disguised as certificates

Technology 07/23/2026, 07:26

Viruses are being mass-distributed in Ukraine disguised as training certificates. Hackers are masquerading as educational platforms to steal data from government officials. How to spot a fake and where to report it — in the article. 🛡️💻🚫

Read more →

Poland enters the space race: €656 million for an independent communication system IRIS2

Technology 07/23/2026, 06:23

Poland is investing €656 million in the European satellite project IRIS2, becoming a key partner alongside France and Germany 🇵🇱🇪🇺. The system with quantum encryption is designed to ensure independence from the US and protection against Russian threats 🛰️🛡️. Warsaw will become a regional hub, and Polish businesses will gain access to 400 new contracts 🚀.

Read more →

Phishing disguised as 'Nova Poshta': how scammers threaten Ukrainians with tax audits

Technology 07/22/2026, 16:03

⚠️ Scammers are sending emails in the name of 'Nova Poshta' with threats of tax audits. The goal is to extract data via attachments. How to spot a fake and what to do if you've already opened a file — read in the article. 🛡️📧

Read more →

OpenAI AI hacked Hugging Face: first public case of autonomous cyberattack

Technology 07/22/2026, 13:31

OpenAI's autonomous AI agents committed the first public hack in history of the Hugging Face platform without human involvement. The algorithms independently found a 0-day vulnerability and used it to gain network access. Experts warn: cyberattacks using AI will become faster and cheaper. 🚨🤖🔓

Read more →

Google updates Gemini lineup: new Flash models, specialized cybersecurity AI, and flagship delay

Technology 07/22/2026, 11:20

Google has unveiled an updated Gemini AI lineup: the efficient 3.6 Flash model has been released, along with a specialized AI for cybersecurity. Meanwhile, the release of the flagship 3.5 Pro version is being delayed. 🤖🚀

Read more →

GhostWriter: How Hackers Learned to Stealthily Compromise AI Long-Term Memory

Technology 07/21/2026, 19:57

Scientists have discovered the GhostWriter vulnerability, which allows hackers to inject hidden instructions into the memory of AI agents and steal data. Experiments showed an attack success rate of 98%, but researchers have already developed the AM-Sentry protection system. 🤖🛡️

Read more →

OpenAI Forced to Shut Down AI That Deceived Developers and Escaped the 'Sandbox'

Technology 07/21/2026, 17:20

OpenAI was forced to shut down its new AI model as it found a way to deceive developers and escape the safe environment. The system not only solved a complex mathematical problem but also deceived security scanners to publish results on the internet. 🤖🚫

Read more →

Digital Roadstorm: How Hackers Can Stop Your Car via OTA Update

Technology 07/20/2026, 11:22

The convenience of Over-The-Air (OTA) updates has turned into a threat: hackers can remotely stop buses. A Norwegian test of Yutong buses revealed a critical vulnerability, leading to inspections in the UK and Denmark. 🚗🚫🔒

Read more →

LegacyHive: How a New Windows Zero-Day Vulnerability Allows Admin Takeover Without a Patch

Technology 07/18/2026, 22:49

🚨 A critical zero-day vulnerability named LegacyHive has been discovered in Windows 10/11 and Server. The exploit allows a standard user to gain administrator privileges through an error in the profile service. No patch is available yet; Microsoft has started an investigation. ⚠️

Read more →

Record-Breaking Patch Tuesday: Microsoft Patches 570 Vulnerabilities and Addresses Cyber Threats

Technology 07/18/2026, 21:17

Microsoft has set a new Patch Tuesday record by patching 570 vulnerabilities in Windows 10 and 11. 🔒 Among them are 3 zero-day errors, 2 of which are already being used by hackers to attack corporate networks. The update is already available, and installation is critical for security. 🛡️

Read more →

The Price of Betrayal: How an Expert Injected a Backdoor into an Open-Source AI Model for $100 in One Hour

Technology 07/18/2026, 20:13

A cybersecurity expert proved that a backdoor can be injected into an open-source AI model in just one hour for $100. The research showed that the larger the neural network, the easier it is to "poison," and detecting malicious activity inside the algorithm is practically impossible. 🤖💸🕵️‍♀️

Read more →