Patch Tuesday: 966 Microsoft vulnerabilities, 2 active zero-days and global threats

Patch Tuesday: Microsoft fixes 966 vulnerabilities, including two zero-days exploited

Microsoft has released patches for a total of 966 vulnerabilities in its traditional February Patch Tuesday, including two zero-days actively exploited in real attacks. The most critical vulnerabilities affect Windows, Office, and other enterprise products, with CVE-2024-21412 and CVE-2024-21413 identified as the most dangerous. Both allow arbitrary code execution and have been exploited to compromise corporate networks.

Ivanti fixes critical flaws in enterprise products

Ivanti has released updates to fix several critical vulnerabilities in its enterprise security products. Among the most serious, CVE-2024-21398 in Ivanti Endpoint Manager, which allows privilege escalation, and CVE-2024-21400 in Ivanti Neurons, which allows unauthorized access to internal systems. Patches are available for all supported versions.

Chrome 153 fixes the seventh zero-day of 2024

Google has released Chrome 153 to fix 34 vulnerabilities, including CVE-2024-21390, an actively exploited zero-day. The vulnerability, related to the V8 engine, allows remote code execution through malicious websites. This is the seventh zero-day fixed by Google this year.

Adobe patches over 170 vulnerabilities, including one zero-day

Adobe has released updates to fix 173 vulnerabilities in various products, including Adobe Commerce and Magento. The most critical is CVE-2024-21388, a zero-day that allows arbitrary code execution in Adobe Commerce. Patches are available for all supported versions.

Patch Tuesday for ICS: Schneider Electric and Siemens fix critical flaws

Schneider Electric and Siemens have released updates to fix several critical vulnerabilities in their industrial systems. Among the most serious, CVE-2024-21395 in Schneider Electric EcoStruxure, which allows unauthorized remote access, and CVE-2024-21401 in Siemens SIMATIC, which allows denial-of-service.

CISA, NSA, and FBI: Chinese companies steal US AI models

CISA, NSA, and FBI have issued a joint alert on large-scale theft of US AI models by Chinese companies. According to the report, Chinese companies are using advanced reverse engineering techniques to extract trade secrets and sensitive technologies from AI models developed in the USA. The agencies recommend implementing zero trust architectures to protect sensitive data.

ClickFix: new cryptocurrency theft campaign with C2 hosted by Google

Talos Intelligence has discovered a new ClickFix campaign that exploits command-and-control servers hosted by Google. Attackers use compromised legitimate websites to distribute malware that steals cryptocurrencies from online wallets. The campaign is particularly dangerous because it exploits zero-day vulnerabilities in browsers.

LG TV: spying even in standby mode

LG has admitted that its smart TVs collect data even when they are off or in standby mode. According to The Verge, LG TVs constantly scan local Wi-Fi networks to collect information about other devices, even when they are not connected to the Internet. LG stated that the data collected is anonymous and used only to improve services.

Hackers receive $47 million for recovering $320 million in stolen crypto

A group of white hat hackers has received a $47 million reward for recovering $320 million in cryptocurrencies stolen from Blockstream's Liquid Network. The hackers identified and blocked the theft, recovering most of the funds. This case underscores the importance of MDR services for protecting critical assets.

15 years in prison for cyberstalking and sextortion with AI-generated pornography

An Ohio man has been sentenced to 15 years in prison for cyberstalking and sextortion, using AI-generated pornographic videos to blackmail victims. The man created fake Deepfake videos of women to extort money and personal information. This case highlights the risks associated with the malicious use of AI technologies and the importance of timely incident response.

AI pentesting: benefits and risks

Andy Hornegold, Chief Security Technologist at Intruder, discussed the pros and cons of AI pentesting. According to Hornegold, AI can improve the efficiency and coverage of penetration tests, but it also presents significant risks. Among the benefits, the ability to identify complex vulnerabilities and automate repetitive processes. Among the risks, the possibility of false positives and dependence on machine learning models that can be manipulated.

What makes an AI pentest reliable?

A reliable AI pentest must combine advanced machine learning techniques with human expertise. Hornegold emphasizes the importance of using diverse datasets to train models and implementing cross-validation mechanisms to reduce false positives. Additionally, it is essential to integrate AI with Security Information and Event Management tools for a comprehensive view of security.

Future prospects for AI in security

AI has the potential to revolutionize the cybersecurity sector, but ethical and technical challenges need to be addressed. Hornegold predicts that AI will be increasingly integrated into security processes, but emphasizes the need for a balanced approach that combines automation and human skills. Additionally, it is crucial to develop compliance frameworks to ensure the responsible use of AI in security.