Cyber OSINT Overview, Sep 1 - Sep 7, 2025 #
This brief consolidates key updates from 80+ sources, including government organizations, cybersecurity vendors, threat intelligence teams, security research labs, and blogs from cybersecurity communities and professionals. It highlights the most significant threats, vulnerabilities, and developments from the past week to keep you informed.
Most Discussed Topics #
- Artificial intelligence (AI) is a dual-use technology in cybersecurity, extensively discussed for both offensive and defensive applications. On the offensive side, AI is used in sophisticated phishing, deepfakes, autonomous ransomware, and malware like DarkBard and PromptLock. Defensively, AI assists in threat detection, vulnerability tracking, AI Security Posture Management (AI-SPM), and LLM-powered threat hunting to combat machine-speed adversaries and manage the growing volume of CVEs.
- community www.reddit.com: Large language models can execute complete ransomware attacks autonomously, research shows
- news cyberscoop.com: AI can help track an ever-growing body of vulnerabilities, CISA official says
- news cyberscoop.com: NYU team behind AI-powered malware dubbed ‘PromptLock’
- news cybersecuritynews.com: Top 10 Best AI Penetration Testing Companies in 2025
- news cybersecuritynews.com: “GPUGate” Malware Abuses Google Ads and GitHub to Deliver Advanced Malware Payload
- news hackread.com: Scammers Exploit Grok AI With Video Ad Scam to Push Malware on X
- news cio.com: AI-driven endpoint security: Staying resilient, everywhere
- news cio.com: Your biggest AI risk might be that employees don’t know they’re using it
- news cio.com: Ransomware ain’t what it used to be
- personal grahamcluley.com: Smashing Security podcast #433: How hackers turned AI into their new henchman
- personal www.schneier.com: Generative AI as a Cybercrime Assistant
- personal www.schneier.com: Indirect Prompt Injection Attacks Against LLM Assistants
- vendor blog.barracuda.com: DarkBard: The “Evil Twin” of Google Bard
- vendor blog.knowbe4.com: CyberheistNews Vol 15 #35 [Watch Out] Hackers Now Use AI to Write Better Phish
- vendor blog.knowbe4.com: Report: AI Can Now Automate Entire Attack Chains
- vendor blog.virustotal.com: Supercharging Your Threat Hunts: Join VirusTotal at Labscon for a Workshop on Automation and LLMs
- vendor www.group-ib.com: From Deepfakes to Dark LLMs: 5 use-cases of how AI is Powering Cybercrime
- Effective vulnerability and patch management is crucial given the continuous stream of newly discovered flaws. Advisories highlight widespread vulnerabilities in critical software and platforms, including Sitecore, Android, Linux Kernel, Cisco, VMware, and SAP. Organizations are urged to apply patches promptly, as evidenced by CISA’s KEV Catalog which mandates rapid remediation for actively exploited vulnerabilities.
- gov advisories.ncsc.nl: NCSC-2025-0245 [1.01] [M/H] Kwetsbaarheden verholpen in SAP producten
- gov advisories.ncsc.nl: NCSC-2025-0273 [1.00] [M/H] Kwetsbaarheden verholpen in Google Android en Samsung Mobile
- gov cyber.gc.ca: Android security advisory – September 2025 monthly rollup (AV25-560)
- gov cyber.gc.ca: Atlassian security advisory (AV25-566)
- gov cyber.gc.ca: Cisco security advisory (AV25-562)
- gov cyber.gc.ca: [Control systems] CISA ICS security advisories (AV25-554)
- gov cyber.gc.ca: Dell security advisory (AV25-558)
- gov cyber.gc.ca: Drupal security advisory (AV25-563)
- gov cyber.gc.ca: Google Chrome security advisory (AV25-561)
- gov cyber.gc.ca: HashiCorp security advisory (AV25-555)
- gov cyber.gc.ca: HPE security advisory (AV25-567)
- gov cyber.gc.ca: IBM security advisory (AV25-557)
- gov cyber.gc.ca: Jenkins security advisory (AV25-564)
- gov cyber.gc.ca: Qualcomm security advisory – September 2025 monthly rollup (AV25-556)
- gov cyber.gc.ca: Red Hat security advisory (AV25-553)
- gov cyber.gc.ca: Sitecore security advisory (AV25-568)
- gov cyber.gc.ca: Ubuntu security advisory (AV25-552)
- gov cyber.gc.ca: VMware security advisory (AV25-565)
- gov cyber.gc.ca: WhatsApp security advisory (AV25-559)
- gov wid.cert-bund.de: [NEU] [mittel] ImageMagick (MagickCore blob): Schwachstelle ermöglicht Codeausführung
- gov www.cisa.gov: CISA Adds Three Known Exploited Vulnerabilities to Catalog
- gov www.cisecurity.org: Multiple Vulnerabilities in Google Chrome Could Allow for Arbitrary Code Execution
- gov www.jpcert.or.jp: Weekly Report: コニカミノルタ製bizhubシリーズにサービス運用妨害(DoS)の脆弱性
- gov www.jpcert.or.jp: Weekly Report: Citrix Netscaler ADCおよびGatewayに脆弱性
- gov www.jpcert.or.jp: Weekly Report: 複数のiND製品に脆弱性
- gov www.jpcert.or.jp: Weekly Report: GitLabに複数の脆弱性
- gov www.jpcert.or.jp: Weekly Report: 複数のi-フィルター製品に不適切なファイルアクセス権設定の脆弱性
- gov www.jpcert.or.jp: Weekly Report: SS1に複数の脆弱性
- gov www.jpcert.or.jp: Weekly Report: ScanSnap Managerのインストーラに権限昇格につながる脆弱性
- gov www.jpcert.or.jp: Weekly Report: Google Chromeに脆弱性
- Supply chain security and managing third-party risk remain critical challenges. Several articles report on the Salesloft Drift integration compromise, which led to mass data exfiltration from numerous customers across multiple integrated services like Salesforce, Google Workspace, and cloud providers. This incident highlights how vulnerabilities in one component can cascade through an organization’s entire digital ecosystem, emphasizing the need for robust third-party risk management and the adoption of tools like Software Bill of Materials (SBOMs) to enhance transparency.
- gov cert.europa.eu: Cyber Brief 25-09 - August 2025
- news www.darkreading.com: Blast Radius of Salesloft Drift Attacks Remains Uncertain
- news www.darkreading.com: Zscaler, Palo Alto Networks Breached via Salesloft Drift
- personal krebsonsecurity.com: The Ongoing Fallout from a Breach at AI Chatbot Maker Salesloft
- vendor cloud.google.com: Widespread Data Theft Targets Salesforce Instances via Salesloft Drift
- vendor fortiguard.fortinet.com: Salesloft Drift Supply Chain Attack
- vendor unit42.paloaltonetworks.com: Model Namespace Reuse: An AI Supply-Chain Attack Exploiting Model Name Trust
- vendor unit42.paloaltonetworks.com: Threat Brief: Salesloft Drift Integration Used To Compromise Salesforce Instances
- vendor www.paloaltonetworks.com: Salesforce-Connected Third-Party Drift Application Incident Response
- vendor www.tenable.com: Tenable Response to Salesforce and Salesloft Drift Incident
- Phishing and social engineering attacks continue to be a primary vector for cybercriminals, with evolving tactics such as smishing (SMS phishing), vishing (voice phishing), and advanced email cloaking techniques. Attackers are increasingly leveraging AI to craft more convincing and tailored messages, making it harder for users to distinguish legitimate communications from malicious ones. Organizations and individuals must maintain high vigilance and implement robust security awareness training.
- community www.reddit.com: GOP Cries Censorship Over Spam Filters That Work
- gov cert.europa.eu: Cyber Brief 25-09 - August 2025
- gov ria.ee: August küberruumis: petturid suvepuhkust ei võtnud
- news hackread.com: Scammers Exploit Grok AI With Video Ad Scam to Push Malware on X
- news www.malwarebytes.com: PayPal users targeted in account profile scam
- news www.malwarebytes.com: Tax refund scam targets Californians
- personal krebsonsecurity.com: GOP Cries Censorship Over Spam Filters That Work
- vendor blog.barracuda.com: Malware Brief: Crafty phishing, BYOVD and Android RATs
- vendor blog.barracuda.com: Threat Spotlight: Tycoon phishing kit reveals new techniques to hide malicious links
- vendor blog.barracuda.com: August’s top threat actors: Ransomware, espionage and infostealers
- vendor blog.knowbe4.com: A Warrant Is Out for Your Arrest
- vendor blog.knowbe4.com: CyberheistNews Vol 15 #35 [Watch Out] Hackers Now Use AI to Write Better Phish
- vendor blog.knowbe4.com: Hospitals Need to Prepare for AI-Powered Phishing Attacks
- vendor blog.knowbe4.com: Smishing Campaign Targets California Taxpayers With Phony Refund Offers
- vendor blog.knowbe4.com: Warning: New Spear Phishing Campaign Targets Executives
- vendor medium.com: Trickbot: Malware Overview
- vendor securelist.com: Cookies and how to bake them: what they are for, associated risks, and what session hijacking has to do with it
- Linux Kernel vulnerabilities are a persistent and high-impact concern, with numerous advisories detailing flaws that can lead to Denial of Service (DoS), privilege escalation, and unspecified system impacts. These vulnerabilities are frequently identified across different Linux distributions, including Red Hat Enterprise Linux and Ubuntu. Continuous patching and system updates are essential to mitigate these pervasive risks.
- gov cyber.gc.ca: Red Hat security advisory (AV25-553)
- gov cyber.gc.ca: Ubuntu security advisory (AV25-552)
- gov wid.cert-bund.de: [UPDATE] [hoch] Linux Kernel: Mehrere Schwachstellen ermöglichen Denial of Service
- gov wid.cert-bund.de: [UPDATE] [mittel] Linux Kernel: Mehrere Schwachstellen ermöglichen Denial of Service
- gov wid.cert-bund.de: [UPDATE] [mittel] Linux Kernel: Mehrere Schwachstellen
- gov wid.cert-bund.de: [UPDATE] [hoch] Linux Kernel: Schwachstelle ermöglicht Denial of Service
- gov wid.cert-bund.de: [UPDATE] [hoch] Linux Kernel: Mehrere Schwachstellen
- gov wid.cert-bund.de: [UPDATE] [hoch] Linux Kernel: Mehrere Schwachstellen
- gov wid.cert-bund.de: [NEU] [hoch] Linux Kernel: Mehrere Schwachstellen ermöglichen Denial of Service
- news thecyberexpress.com: Google Patches 111 Android Vulnerabilities, Confirms Active Exploitation of Two Zero-Days
- news thecyberexpress.com: Critical Linux UDisks Daemon Vulnerability (CVE-2025-8067) Exposes Privileged Data to Local Attackers
Critical Vulnerabilities #
- An actively exploited critical vulnerability (CVE-2025-53690) in Sitecore Experience Manager, Platform, and Commerce allows remote code execution due to the use of publicly exposed sample machine keys. This flaw can lead to initial server compromise, privilege escalation, and full system control. CISA has added this to its Known Exploited Vulnerabilities (KEV) Catalog, mandating federal agencies to remediate it immediately.
- gov advisories.ncsc.nl: NCSC-2025-0245 [1.01] [M/H] Kwetsbaarheden verholpen in SAP producten
- gov cyber.gc.ca: Sitecore security advisory (AV25-568)
- gov wid.cert-bund.de: [UPDATE] [kritisch] Sitecore Experience Manager: Mehrere Schwachstellen
- gov wid.cert-bund.de: [NEU] [kritisch] Sitecore Experience Manager: Schwachstelle ermöglicht Codeausführung
- gov www.cisa.gov: CISA Adds Three Known Exploited Vulnerabilities to Catalog
- news cyberscoop.com: Sitecore zero-day vulnerability springs up from exposed machine key
- news therecord.media: CISA orders federal agencies to patch Sitecore zero-day following hacking reports
- news www.darkreading.com: Sitecore Zero-Day Sparks New Round of ViewState Threats
- vendor cloud.google.com: ViewState Deserialization Zero-Day Vulnerability in Sitecore Products (CVE-2025-53690)
- Google’s September 2025 Android Security Bulletin includes patches for two actively exploited zero-day vulnerabilities: CVE-2025-38352 (Linux Kernel Time-of-Check Time-of-Use race condition) and CVE-2025-48543 (Android Runtime unspecified vulnerability). Both are high-severity Elevation of Privilege (EoP) flaws that do not require user interaction for exploitation. Users are strongly advised to apply the necessary updates to their Android devices immediately.
- gov advisories.ncsc.nl: NCSC-2025-0273 [1.00] [M/H] Kwetsbaarheden verholpen in Google Android en Samsung Mobile
- gov cyber.gc.ca: Android security advisory – September 2025 monthly rollup (AV25-560)
- gov wid.cert-bund.de: [UPDATE] [hoch] Android Patchday März 2025: Mehrere Schwachstellen
- gov wid.cert-bund.de: [UPDATE] [hoch] Android Patchday April 2025: Mehrere Schwachstellen
- gov wid.cert-bund.de: [UPDATE] [hoch] Android Patchday Mai 2025: Mehrere Schwachstellen
- gov wid.cert-bund.de: [UPDATE] [hoch] Android Patchday Juni 2025: Mehrere Schwachstellen
- gov wid.cert-bund.de: [UPDATE] [hoch] Android Patchday September 2025: Multiple Vulnerabilities
- gov wid.cert-bund.de: [UPDATE] [hoch] Pixel Patchday September 2025: Mehrere Schwachstellen
- gov www.cisa.gov: CISA Adds Three Known Exploited Vulnerabilities to Catalog
- news cyberscoop.com: Google patches two Android zero-days, 120 defects total in September security update
- news thecyberexpress.com: Google Patches 111 Android Vulnerabilities, Confirms Active Exploitation of Two Zero-Days
- news www.malwarebytes.com: Update your Android! Google patches 111 vulnerabilities, 2 are critical
- A zero-click authorization vulnerability (CVE-2025-55177) in WhatsApp for iOS and Mac, in conjunction with an Apple OS-level flaw (CVE-2025-43300) in the ImageIO framework, has been actively exploited in targeted attacks. This allowed remote code execution by processing malicious content from arbitrary URLs via abused linked device synchronization messages. WhatsApp and Apple have released patches, urging users to update their devices and perform factory resets if compromised.
- gov cyber.gc.ca: WhatsApp security advisory (AV25-559)
- gov www.cisa.gov: CISA Adds Two Known Exploited Vulnerabilities to Catalog
- news www.darkreading.com: WhatsApp Bug Anchors Targeted Zero-Click iPhone Attacks
- news www.malwarebytes.com: WhatsApp fixes vulnerability used in zero-click attacks
- vendor blog.quarkslab.com: Reverse engineering of Apple's iOS 0-click CVE-2025-43300: 2 bytes that make size matter
- vendor research.checkpoint.com: 1st September – Threat Intelligence Report
- Two actively exploited vulnerabilities in end-of-life TP-Link routers (Archer C7 and TL-WR841N/ND), CVE-2023-50224 (authentication bypass/password theft) and CVE-2025-9377 (OS command injection), are being chained by the Quad7 botnet. This botnet weaponizes infected routers to launch widespread password spraying attacks against Microsoft 365 accounts. CISA has added these to its KEV catalog, urging users to update firmware or replace devices.
- gov www.cisa.gov: CISA Adds Two Known Exploited Vulnerabilities to Catalog
- news gbhackers.com: CISA Warns: TP-Link Vulnerabilities Under Active Exploitation
- news www.malwarebytes.com: TP-Link warns of botnet infecting routers and targeting Microsoft 365 accounts
- Multiple high-severity vulnerabilities, including a memory buffer overflow (CVE-2025-2521) and an integer underflow (CVE-2025-2523), have been identified in Honeywell OneWireless Wireless Device Manager (WDM). These remotely exploitable flaws, with low attack complexity, could lead to remote code execution, information exposure, or denial of service. Users and administrators are urged to review advisories and apply necessary updates to affected versions.
- gov www.cisa.gov: CISA Releases Five Industrial Control Systems Advisories
- gov www.cisa.gov: Honeywell OneWireless Wireless Device Manager (WDM)
- A critical and actively exploited code injection vulnerability (CVE-2025-42957) exists in SAP S/4HANA, SAP NetWeaver Application Server ABAP, and other related SAP products. This flaw, with a CVSS score of 9.9, allows a low-privileged user to execute arbitrary code and gain full control of the SAP system and its host OS. SAP released patches in August 2025, which organizations must apply immediately to prevent system compromise.
- gov advisories.ncsc.nl: NCSC-2025-0245 [1.01] [M/H] Kwetsbaarheden verholpen in SAP producten
- gov wid.cert-bund.de: [UPDATE] [mittel] Apache Kafka: Mehrere Schwachstellen
- news gbhackers.com: Critical SAP S/4HANA Vulnerability Actively Exploited, Allowing Full System Takeover
- news hackread.com: Critical SAP Vulnerability CVE-2025-42957 Actively Exploited by Hackers
- news www.darkreading.com: Critical SAP S/4HANA Vulnerability Under Attack, Patch Now
- vendor securelist.com: IT threat evolution in Q2 2025. Non-mobile statistics
- A critical security flaw (CVE-2025-8067) in the Linux UDisks daemon, an out-of-bounds read vulnerability, allows local, unprivileged users to access files and data owned by privileged accounts. Red Hat classified this flaw as ‘Important’ with a CVSS v3.1 base score of 8.5. This can lead to system crashes (DoS) or the disclosure of sensitive memory, including cryptographic keys and PII.
- gov wid.cert-bund.de: [UPDATE] [hoch] Linux UDisks Daemon: Schwachstelle ermöglicht Privilegieneskalation
- news thecyberexpress.com: Critical Linux UDisks Daemon Vulnerability (CVE-2025-8067) Exposes Privileged Data to Local Attackers
- Multiple vulnerabilities are reported in widely used software and hardware. These include a local escalation of privilege in HPE M-Series Switches, various flaws in Atlassian’s Bamboo, Bitbucket, and Crowd, and vulnerabilities in VMware Tanzu products. Cisco also released advisories for numerous phone, network, and collaboration products. Google Chrome, IBM, Dell, and Qualcomm also released security updates for their respective products. Drupal’s Acquia DAM had an access bypass/information disclosure flaw. Jenkins plugins and HashiCorp Vault also required security updates, including a DoS vulnerability in Vault.
- gov cyber.gc.ca: Atlassian security advisory (AV25-566)
- gov cyber.gc.ca: Cisco security advisory (AV25-562)
- gov cyber.gc.ca: Dell security advisory (AV25-558)
- gov cyber.gc.ca: Drupal security advisory (AV25-563)
- gov cyber.gc.ca: Google Chrome security advisory (AV25-561)
- gov cyber.gc.ca: HashiCorp security advisory (AV25-555)
- gov cyber.gc.ca: HPE security advisory (AV25-567)
- gov cyber.gc.ca: IBM security advisory (AV25-557)
- gov cyber.gc.ca: Jenkins security advisory (AV25-564)
- gov cyber.gc.ca: Qualcomm security advisory – September 2025 monthly rollup (AV25-556)
- gov cyber.gc.ca: VMware security advisory (AV25-565)
- Industrial Control Systems (ICS) and Operational Technology (OT) environments face ongoing critical vulnerabilities. CISA released advisories for various products, including Delta Electronics EIP Builder (XML External Entity Reference, CVE-2025-57704), Fuji Electric FRENIC-Loader 4 (Deserialization of Untrusted Data, CVE-2025-9365), and SunPower PVS6 (Hard-Coded Credentials, CVE-2025-9696). These flaws can lead to information disclosure, arbitrary code execution, and full device access. Users are advised to apply updates and implement robust network segmentation and secure remote access.
- community isc.sans.edu: Exploit Attempts for Dassault DELMIA Apriso. CVE-2025-5086, (Wed, Sep 3rd)
- gov cyber.gc.ca: [Control systems] CISA ICS security advisories (AV25-554)
- gov www.cisa.gov: CISA Releases Four Industrial Control Systems Advisories
- gov www.cisa.gov: CISA Releases Five Industrial Control Systems Advisories
- gov www.cisa.gov: Delta Electronics EIP Builder
- gov www.cisa.gov: Fuji Electric FRENIC-Loader 4
- gov www.cisa.gov: SunPower PVS6
Major Incidents #
- A widespread supply chain attack targeted Salesloft’s Drift AI chatbot integration, leading to mass exfiltration of OAuth tokens and sensitive data from hundreds of customer Salesforce instances and integrated services. This impacted major organizations including Google, Cloudflare, Zscaler, Palo Alto Networks, and Tenable. Attackers, tracked as UNC6395, actively deleted forensic evidence after obtaining credentials like AWS keys, Snowflake tokens, VPN credentials, and API keys.
- gov cert.europa.eu: Cyber Brief 25-09 - August 2025
- personal krebsonsecurity.com: The Ongoing Fallout from a Breach at AI Chatbot Maker Salesloft
- personal www.cybersecuritypulse.net: TCP #101: Salesloft Breach; Cato Acquires Aim; Earnings; and Product News
- vendor fortiguard.fortinet.com: Salesloft Drift Supply Chain Attack
- vendor unit42.paloaltonetworks.com: Threat Brief: Salesloft Drift Integration Used To Compromise Salesforce Instances
- vendor www.paloaltonetworks.com: Salesforce-Connected Third-Party Drift Application Incident Response
- vendor www.tenable.com: Tenable Response to Salesforce and Salesloft Drift Incident
- Luxury automaker Jaguar Land Rover (JLR) experienced a cyberattack that severely disrupted its global IT systems, leading to factory closures and significant operational halts. A group identified as ‘Scattered Lapsus$ Hunters’ claimed responsibility for the incident, asserting data exfiltration. The UK Information Commissioner’s Office confirmed a data breach, prompting JLR to proactively take down systems to mitigate impact.
- community www.reddit.com: Disruption to Jaguar Land Rover after cyber-attack may last until October
- gov ctoatncsc.substack.com: CTO at NCSC Summary: week ending September 7th
- news cybersecuritynews.com: Weekly Cybersecurity News Recap : Palo Alto Networks, Zscaler, Jaguar Land Rover, and Cyber Attacks
- news hackread.com: Bridgestone Confirms Cyberattack Disrupting North American Plants
- news therecord.media: Cyberattack forces Jaguar Land Rover to tell staff to stay at home
- news www.darkreading.com: Jaguar Land Rover Shuts Down in Scramble to Secure 'Cyber Incident'
- The Texas Attorney General has filed a lawsuit against PowerSchool, an education technology provider, following a data breach in December 2024. This incident exposed sensitive personal information of over 880,000 Texas school children and teachers, including names, addresses, Social Security Numbers, and medical details. The breach originated from a subcontractor’s account that allegedly lacked adequate security protections.
- news thecyberexpress.com: 880,000 Texans Affected in PowerSchool Data Breach; Attorney General Launches Legal Action
- American consumer credit reporting agency TransUnion suffered a data breach resulting in the exposure of sensitive personal information for over 4.4 million individuals in the United States. The compromised data included names, billing addresses, phone numbers, email addresses, dates of birth, unredacted Social Security Numbers, transaction reasons, and customer support messages. This highlights the ongoing risks associated with large-scale data custodians.
- vendor research.checkpoint.com: 1st September – Threat Intelligence Report
- Tire manufacturing giant Bridgestone Americas confirmed a cyberattack that disrupted some of its North American manufacturing facilities, leading to operational pauses for employees. While the company stated the incident was contained and no customer data was compromised, the nature of the attack is under investigation. The ‘Scattered Lapsus$ Hunters’ group, known for data theft and extortion, is suspected to be involved.
- news hackread.com: Bridgestone Confirms Cyberattack Disrupting North American Plants
- news www.darkreading.com: Bridgestone Americas Confirms Cyberattack
- Chess.com, a major online chess platform, experienced a limited data breach affecting 4,541 users, including residents of Maine and Vermont. The breach occurred in June 2025 through a compromised third-party file transfer tool, exposing names and other non-sensitive identifiers. No passwords or financial information were affected, and federal law enforcement agencies are involved in the ongoing investigation.
- news hackread.com: Chess.com Hit by Limited Data Breach Linked to 3rd-Party File Transfer Tool
- news thecyberexpress.com: Chess.com Confirms Data Breach Exposing Information of 4,541 Users
- A major cyberattack on Miljödata, a Swedish IT provider, disrupted services for over 200 municipalities. The incident raised significant concerns regarding the potential theft of sensitive personal data, including medical certificates, rehabilitation cases, and occupational injury reports. This highlights the critical impact on public services and sensitive data when core IT infrastructure providers are compromised.
- vendor research.checkpoint.com: 1st September – Threat Intelligence Report
Emerging Threats #
- Cybercriminals are increasingly leveraging AI to automate and scale offensive operations. Research demonstrates that Large Language Models (LLMs) can autonomously plan, adapt, and execute ransomware attacks, performing reconnaissance, credential harvesting, network penetration, and generating personalized extortion demands and notes. New malicious AI tools like ‘DarkBard,’ an ’evil twin’ to Google Bard, are also marketed for misinformation, deepfakes, and sophisticated phishing campaigns, enabling a single attacker to operate like an entire cybercrime team.
- community www.reddit.com: Large language models can execute complete ransomware attacks autonomously, research shows
- news cyberscoop.com: NYU team behind AI-powered malware dubbed ‘PromptLock’
- news cio.com: Ransomware ain’t what it used to be
- personal grahamcluley.com: Smashing Security podcast #433: How hackers turned AI into their new henchman
- personal www.schneier.com: Generative AI as a Cybercrime Assistant
- personal www.schneier.com: Indirect Prompt Injection Attacks Against LLM Assistants
- vendor blog.barracuda.com: DarkBard: The “Evil Twin” of Google Bard
- vendor blog.knowbe4.com: Report: AI Can Now Automate Entire Attack Chains
- vendor www.group-ib.com: From Deepfakes to Dark LLMs: 5 use-cases of how AI is Powering Cybercrime
- A sophisticated malvertising campaign, dubbed ‘GPUGate,’ is leveraging Google Ads and manipulated GitHub repositories to trick users into downloading malicious software. This malware employs a unique GPU-gated decryption routine that only activates on systems with a genuine Graphics Processing Unit (GPU), designed to evade traditional security sandboxes. The campaign targets IT professionals in Western Europe, with evidence suggesting Russian-speaking threat actors are behind it.
- The Russian state-sponsored hacking group APT28 (Fancy Bear) is deploying a new backdoor, ‘NotDoor,’ by abusing Microsoft Outlook. This malware operates as a Visual Basic for Applications (VBA) macro within Outlook, activated by specific email trigger phrases. It is deployed via DLL sideloading of a malicious ‘SSPICLI.dll’ through Microsoft’s signed OneDrive.exe to disable macro security and exfiltrate data covertly. This tactic blends with trusted binaries and normal mail flow, bypassing basic perimeter tools.
- news hackread.com: Russian APT28 Deploys “NotDoor” Backdoor Through Microsoft Outlook
- news www.darkreading.com: Russia's APT28 Targets Microsoft Outlook With 'NotDoor' Malware
- The Quad7 botnet is actively exploiting two vulnerabilities (CVE-2023-50224 for password theft and CVE-2025-9377 for OS command injection) in end-of-life TP-Link Archer C7 and TL-WR841N/ND routers. This botnet weaponizes infected routers by turning them into nodes for widespread password spraying attacks against Microsoft 365 accounts. The distributed nature of the botnet, using thousands of residential IP addresses, makes detection particularly challenging for defenders.
- gov www.cisa.gov: CISA Adds Two Known Exploited Vulnerabilities to Catalog
- news gbhackers.com: CISA Warns: TP-Link Vulnerabilities Under Active Exploitation
- news www.malwarebytes.com: TP-Link warns of botnet infecting routers and targeting Microsoft 365 accounts
- A new, potentially China-based, threat actor group dubbed ‘GhostRedirector’ is compromising Windows servers with Rungan and Gamshen backdoors. Gamshen is a malicious IIS module designed to manipulate Google search rankings, driving traffic to gambling websites through an ‘SEO fraud-as-a-service’ scheme. This operation has affected at least 65 Windows servers globally across various industries, emphasizing opportunistic attacks rather than specific targeting.
- news therecord.media: 'SEO fraud-as-a-service' scheme hijacks Windows servers to promote gambling websites
- news www.darkreading.com: Chinese Hackers Game Google to Boost Gambling Sites
- vendor www.welivesecurity.com: GhostRedirector poisons Windows servers: Backdoors with a side of Potatoes
- The secretive Malware-as-a-Service (MaaS) group TAG-150 is rapidly developing and deploying self-developed malware families, including CastleLoader, CastleBot, and the newly documented remote access trojan (RAT) CastleRAT. Initial infections often stem from Cloudflare-themed ‘ClickFix’ phishing attacks or fraudulent GitHub repositories. This group operates a multi-tiered infrastructure with sophisticated command-and-control (C2) mechanisms, showing advanced operational security and redundancy planning.
- news cybersecuritynews.com: TAG-150 Hackers Deploying Self-Developed Malware Families to Attack Organizations
- news www.darkreading.com: Secretive MaaS Group 'TAG-150' Develops Novel 'CastleRAT'
- vendor www.recordedfuture.com: From CastleLoader to CastleRAT: TAG-150 Advances Operations with Multi-Tiered Infrastructure
- North Korea-aligned threat actors are actively monitoring cyber threat intelligence (CTI) platforms like Validin, VirusTotal, and Maltrail to assess their infrastructure exposure and scout for new assets. These actors operate in coordinated teams, demonstrate real-time collaboration, and rapidly deploy new infrastructure in response to service provider takedowns, indicating a strategic focus on sustaining operations and high victim engagement rather than preventing detection.
- vendor www.sentinelone.com: Contagious Interview | North Korean Threat Actors Reveal Plans and Ops by Abusing Cyber Intel Platforms
Regulatory and Policy Updates #
- CISA, in collaboration with the NSA and 19 international partners, released joint guidance on ‘A Shared Vision of Software Bill of Materials (SBOM) for Cybersecurity.’ This initiative aims to strengthen software supply chain transparency, improve vulnerability management, and enhance global cybersecurity resilience through standardized SBOM adoption. SBOMs act as an ‘ingredients list’ for software, providing essential visibility into dependencies to identify risks and proactively mitigate vulnerabilities.
- A U.S. House panel advanced legislation to reauthorize the 2015 Cybersecurity and Information Sharing Act and the State and Local Cybersecurity Grant Program, both nearing expiration. These bills, known as the WIMWIG Act and the PILLAR Act respectively, aim to update legal definitions for hacking tactics, preserve privacy protections, and encourage the use of secure AI, strengthening cybersecurity for state and local governments and critical infrastructure.
- The Federal Trade Commission (FTC) announced a settlement with Chinese robot toy manufacturer Apitor Technology for illegally collecting geolocation data of US children via its companion app, violating the Children’s Online Privacy Protection Act (COPPA). Apitor is required to pay a $500,000 civil fine, delete illegally collected data, obtain explicit parental consent, and undergo ten years of compliance monitoring. This emphasizes stringent enforcement of children’s online privacy regulations.
- news cyberscoop.com: FTC announces settlement with toy robot makers that tracked location of children
- personal www.bitdefender.com: Parents warned that robot toys spied on children’s location without consent
- A regional cyber alliance has been formed by Ukraine, Romania, and Moldova to enhance cooperation against cyber and hybrid threats, particularly those from Russia, aiming to strengthen collective cyber defense. Concurrently, Romania has established a dedicated Computer Security Incident Response Team (CSIRT) for its energy sector, underscoring efforts to bolster regional energy security in the face of escalating geopolitical cyber risks.
- gov cert.europa.eu: Cyber Brief 25-09 - August 2025
- The UK Department for Science, Innovation and Technology is consulting on proposals to update the Telecommunications Security Code of Practice 2022. These updates are intended to help public telecoms providers protect UK telecommunications networks and services against evolving threats and emerging technologies. This proactive measure aims to enhance national infrastructure resilience in a dynamic threat landscape.
- gov ctoatncsc.substack.com: CTO at NCSC Summary: week ending September 7th
- Czechia’s National Cyber and Information Security Agency (NÚKIB) issued a ‘High’ threat warning concerning Chinese data transfers and remote administration of technical assets from China. The advisory urges both government bodies and private businesses in Czechia to bolster their defenses against state-sponsored espionage campaigns, citing China’s national security laws that compel data assistance to state authorities, potentially compromising data security for critical infrastructure.
- news thecyberexpress.com: Czechia Warns of Chinese Data Transfers and Remote Administration for Espionage
- news www.darkreading.com: Czech Warning Highlights China Stealing User Data
- France’s data watchdog, the CNIL, fined Google €325 million (approximately $381 million) for inserting promotional messages directly into Gmail inboxes without prior consent and for cookie consent practices deemed coercive. This action, part of Europe’s stricter privacy framework (GDPR), highlights an ongoing friction between Google’s advertising business model and regulatory demands for explicit, unambiguous user consent regarding data collection and targeted advertising.
- news thecyberexpress.com: Google Slapped with $381 Million Fine in France Over Gmail Ads, Cookie Consent Missteps
Security Operations #
- Robust incident response capabilities and advanced XDR (Extended Detection and Response) platforms are essential for effective cybersecurity. These platforms provide rapid detection, verification, containment, and eradication of threats by integrating telemetry from endpoints, networks, emails, and identity systems. Continuous monitoring and a multi-tool approach, as demonstrated at Black Hat, are crucial for triaging incidents and swiftly responding to advanced threats.
- news cybersecuritynews.com: Weekly Cybersecurity News Recap : Palo Alto Networks, Zscaler, Jaguar Land Rover, and Cyber Attacks
- news thecyberexpress.com: Black Hat USA 2025 CISO Podcast Series Episode 5 Out Now
- vendor arcticwolf.com: Understanding the Role of Incident Response in Cybersecurity
- vendor blogs.cisco.com: Black Hat USA 2025: 10 Years Protecting Black Hat
- vendor blogs.cisco.com: Black Hat Training Attendees Scan Aviation Organization
- vendor blogs.cisco.com: Conference Hopping: Training Attendee Scanning Def Con
- vendor blogs.cisco.com: Driving Cisco XDR Integration With Third-Party Partners at Black Hat
- vendor www.security.com: Understanding the Role of Incident Response in Cybersecurity
- vendor www.security.com: Why Your Defense in Depth Plan Might Be All Defense and No Depth
- AI and Large Language Models (LLMs) are transforming threat hunting and vulnerability management by automating traditionally manual processes. Tools like VirusTotal’s Code Insight use LLMs for malware analysis across diverse file formats (SWF, SVG), summarizing suspicious logic and identifying campaigns. CISA also highlights AI’s potential to track the rapidly expanding CVE catalog, enabling faster identification of anomalies and more proactive threat responses.
- news cyberscoop.com: AI can help track an ever-growing body of vulnerabilities, CISA official says
- vendor blog.virustotal.com: Supercharging Your Threat Hunts: Join VirusTotal at Labscon for a Workshop on Automation and LLMs
- vendor blog.virustotal.com: Uncovering a Colombian Malware Campaign with AI Code Analysis
- vendor medium.com: August’s Updates: Fresh Connectors, SDK Update, and 2,200+ New Detection Rules
- vendor www.recordedfuture.com: The Future of Cyber Defense Is Autonomous — And It's Coming to Predict 2025
- Zero Trust Architecture (ZTA) and Automated Moving Target Defense (AMTD) are critical for modern security strategies, especially in hybrid and cloud environments. ZTA principles like continuous verification and adaptive access minimize attack surfaces, while AMTD dynamically shifts the attack surface to disrupt reconnaissance and proactively neutralize threats before execution. This prevention-first approach strengthens endpoint protection and enhances overall resilience against sophisticated, fileless, and in-memory attacks.
- vendor www.morphisec.com: How Exposure Management Powers Preemptive Cyber Defense
- vendor www.morphisec.com: Why Automated Moving Target Defense Is Emerging as a Zero Trust Essential
- vendor www.security.com: At DEF CON 33, the Power of Responsible Disclosure Takes Center Stage
- vendor www.security.com: Why Your Defense in Depth Plan Might Be All Defense and No Depth
- Software Bill of Materials (SBOM) are being promoted as a fundamental tool for proactive cybersecurity defense and software supply chain transparency. Joint guidance from CISA, NSA, and international partners highlights SBOMs’ role in providing critical visibility into software dependencies, improving vulnerability management, and supporting secure-by-design principles. Widespread adoption is seen as essential for enhancing resilience and measurably reducing risks and costs across the global software ecosystem.
- Consistent and up-to-date cyber hygiene training is crucial for mitigating human risk, particularly against insider threats and sophisticated social engineering attacks like phishing. Organizations, including educational institutions, must prioritize revisiting training programs to address known weaknesses and emerging digital threats. This emphasis on security awareness training is proven to reduce insider threats and represents a cost-effective investment in overall cybersecurity posture.
- community www.reddit.com: Free Interactive 3D Security Awareness Exercises (Better Alternative to Boring Yearly Training)
- vendor blog.barracuda.com: Back to school, back to scams part 2: Mitigation in motion
- vendor blog.knowbe4.com: Beyond the Audit Box: Building Security That Works in the Real World
- Enhanced packet-level analysis is transforming firewall investigations in Security Operations Centers (SOCs). The integration of advanced packet capture (PCAP) platforms, such as Endace Vision with Cisco’s Firepower Management Center, enables SOC analysts to instantly access definitive network traffic data. This capability drastically reduces the time for forensic analysis, accelerates root cause identification, and allows for more precise and rapid containment of security incidents, turning hours of manual effort into moments.
- vendor blogs.cisco.com: The Value of PCAP in Firewall Investigations
- Organizations using Amazon Web Services (AWS) should implement a comprehensive cloud security strategy that combines native AWS tools with Cloud-Native Application Protection Platforms (CNAPPs). Native tools cover identity (IAM, Access Analyzer), network security (Security Groups, WAF), and data protection (KMS, Macie). CNAPPs augment these by providing multi-cloud support, identity-centric security, and unified risk reduction, correlating identity data with runtime behavior and asset sensitivity to uncover ’toxic combinations’ and reduce overall risk.
- vendor www.tenable.com: Building a Cloud Security Strategy with AWS Native Tools
- vendor www.tenable.com: Identity-First Security: Mitigating the Cloud’s Greatest Risk Vector
Wins #
- Egyptian authorities, in collaboration with the Alliance for Creativity and Entertainment (ACE), successfully dismantled Streameast, the world’s largest illicit live sports streaming network. This operation led to the arrest of two operators in Egypt and the seizure of infrastructure and significant funds. Streameast had amassed over 1.6 billion visits across 80 domains, impacting sports leagues and entertainment companies globally.
- news cyberscoop.com: Streameast, world’s largest pirated live sports network, shut down by Egyptian authorities
- news thecyberexpress.com: Streameast Piracy Network Dismantled After Global Enforcement Operation
- A significant step in international cyber cooperation has been achieved with the formation of a regional cyber alliance between Ukraine, Romania, and Moldova. This alliance aims to bolster collective cyber defense and enhance cooperation against cyber and hybrid threats, particularly from Russia. Furthermore, Romania has established a dedicated Computer Security Incident Response Team (CSIRT) for its energy sector, strengthening regional energy security.
- gov cert.europa.eu: Cyber Brief 25-09 - August 2025
- The U.S. Department of Justice (DOJ) successfully initiated a civil forfeiture action to seize $848,247 in Tether (USDT). These funds were identified as proceeds from elaborate cryptocurrency confidence scams that defrauded victims across several states. This action highlights ongoing efforts by law enforcement to combat crypto-related fraud and recover stolen assets.
- news thecyberexpress.com: DOJ Moves to Seize $848K in Tether Linked to Crypto Confidence Scams
- The RapperBot DDoS botnet, which notably leveraged DNS TXT records for covert command-and-control (C2) communications and employed multi-architecture payloads, has been successfully disrupted. This takedown was a result of law enforcement efforts, specifically the DOJ’s Operation PowerOFF. The success demonstrates effective collaboration in combating sophisticated botnet operations.
- community www.reddit.com: RapperBot: infection → DDoS in seconds (deep dive write-up)
- Panoptic, with support from Cantina and SEAL911, conducted a successful whitehat rescue in the DeFi space. This operation highlights the effectiveness of collaborative efforts within bug bounty programs to identify and mitigate vulnerabilities, thereby strengthening the security and resilience of blockchain and decentralized finance ecosystems.
- personal newsletter.blockthreat.io: BlockThreat - Week 35, 2025
- A hacker in Spain was arrested for accessing a government website to alter high school and university entrance exam grades for himself and classmates. This swift law enforcement action underscores the importance of securing educational systems and holding individuals accountable for exploiting digital platforms for fraudulent purposes.
- personal www.bitdefender.com: Hacker suspected of trying to cheat his way into university is arrested in Spain
Disclaimer #
The summaries in this brief are generated autonomously by a LLM based on provided system and user prompts. While every effort is made to consolidate accurate and relevant insights, the model may occasionally misinterpret, misrepresent, or hallucinate information. Readers are strongly advised to verify all key points by consulting the original sources linked in the brief for complete context and accuracy.
The brief is created in collaboration with BlackStork and is based on a free template.
Reach out if you have questions or suggestions.