DeepSeek on DanSec Daily

Cybersecurity Brief – 2026-09-12

Sat, 12 Sep 2026 00:00:00 +0000

Major Incidents or Breaches

The Florida Department of Highway Safety and Motor Vehicles confirmed a data breach of its DAVID driver database, with attackers accessing the system using stolen police credentials ^[11].
Hackers compromised the Brevo email marketing platform, resulting in phishing emails being sent to at least 347,000 users of Trezor, BitBox, and CoinTracking. Other customers of Brevo may also be at risk ^[30]^[31].

Newly Discovered Vulnerabilities

GitLab disclosed and patched a critical CVSS 10 vulnerability (path traversal) that allows unauthenticated attackers to read arbitrary files from GitLab servers. The flaw was exploited in the wild within hours of public disclosure ^[6]^[28].
Critical and high-severity vulnerabilities in JFrog Artifactory are being exploited in chained attacks, enabling threat actors to bypass authentication, gain administrative privileges, and deploy Rust-based backdoor malware on vulnerable self-hosted servers ^[13].
The BlueMoon exploit kit is chaining recent Chrome and Windows zero-day vulnerabilities in attacks attributed to multiple espionage-motivated threat actors ^[25].

Notable Threat Actor Activity

Multiple threat groups, including financially motivated and state-sponsored actors from Russia and China, have abused Anthropic’s Claude AI model for a range of malicious purposes, including malware development, data theft, and attempts to extract secrets from 1.8 million Android apps ^[8]^[9]^[10]^[14]^[22].
Anthropic identified and disrupted industrial-scale illicit distillation attacks against its Claude AI model by seven China-based AI labs, including entities linked to Alibaba, Moonshot, DeepSeek, and Z.ai ^[7].
Russian state-sponsored hackers used Claude to develop workflows for rebuilding malware after detection ^[9].
Threat actors linked to ShinyHunters, Helix, and other extortion groups are using passkey and single sign-on-themed phishing attacks to compromise corporate Microsoft 365 accounts and steal data ^[12].
An attacker used a semi-autonomous coding agent to harvest and resell LLM API access by exploiting poorly secured resale gateways and web flaws ^[24].
The Gigabud Android banking Trojan creates a hidden copy of a victim’s banking app in a separate work profile to conceal fraudulent transactions ^[32].

Trends, Tools, or Tactics of Interest

OAuth consent phishing attacks are targeting users of messaging apps, with the FBI issuing an advisory about campaigns focused on prominent individuals and their associates ^[1].
Phishing campaigns are increasingly using SVG image files to deliver malicious JavaScript payloads to employees ^[3].
There is a surge in AI-enabled phishing and fraud, with 60 percent of companies reporting increased losses and AI-generated phishing being the top concern ^[2]^[15]^[17].
Threat actors are leveraging trusted AI platforms to host malicious content, poison search results, and distribute malware, increasing the attack surface for AI users ^[14].
The use of AI agents and swarm attacks is altering the cyber kill chain, with attackers employing AI for staging, reconnaissance, lateral movement, and exfiltration ^[19].
Invisible Unicode characters are being used to bypass phishing filters in email attacks ^[29].
A major phishing campaign generated 1 million personalized fraud emails in three days using AI ^[15].
Passkey and single sign-on-themed phishing attacks are being used to target Microsoft 365 users ^[12].
Analysis of simulated phishing attacks suggests organizations should focus on measuring credential leaks and reporting rates, not just click rates, for security awareness ^[27].

Regulatory or Policy Developments Affecting the Security Industry

A new joint government advisory from CISA signals a regulatory shift, calling for more transparent breach notification and incident response protocols ^[16].
Meta is facing a proposed class action lawsuit alleging illegal harvesting of Facebook and Instagram photos to train AI image-generation and face recognition models ^[23].

Sources

Florida confirms DMV database breached via stolen police account - bleepingcomputer.com
Trezor Says 347,000 Users Received Phishing Emails After Brevo Hack - securityweek.com
Crypto customers targeted by scammers after email marketing provider breach - malwarebytes.com
GitLab CVSS 10 File-Read Flaw Draws In-the-Wild Probes After Disclosure - thehackernews.com
GitLab Vulnerability Exploited One Day After Disclosure - securityweek.com
Artifactory flaws chained in attacks deploying backdoor malware - bleepingcomputer.com
BlueMoon Exploit Kit Chains Recent Chrome, Windows Zero-Days - securityweek.com
Claude Used to Automate Exploitation and Data Theft Across Multiple Victims - thehackernews.com
Russian State-Sponsored Hackers Use Claude to Rebuild Malware After Detection - thehackernews.com
Hackers abused Claude to extract secrets from 1.8M Android apps - bleepingcomputer.com
How Threat Actors Are Turning Trusted AI Platforms Into an Attack Surface - bleepingcomputer.com
From Hacks to Bioweapons, Claude Misuse Is Now Everywhere - wired.com
Anthropic Says Seven China-Based AI Labs Ran Industrial-Scale Claude Distillation Attacks - thehackernews.com
Passkey-themed phishing attacks lead to Microsoft 365 data theft - bleepingcomputer.com
The Self-Expanding Stolen Inference Supply Chain: An AI Agent Harvesting and Re-Serving LLM Access, (Fri, Sep 11th) - isc.sans.edu
Android malware creates a hidden copy of your banking app - malwarebytes.com
FBI Alert: OAuth Consent Phishing is Targeting Users of Messaging Apps - blog.knowbe4.com
Phishing Campaign Targets Employees with Malicious SVG Files - blog.knowbe4.com
Survey: Companies Cite Phishing as their Top AI-Enabled Fraud Concern - blog.knowbe4.com
Threat Actor Generates 1M Personalized Fraud Emails in 3 Days - darkreading.com
Why AI Is So Good at Scamming Humans - darkreading.com
Papercut AI Swarm Attack Heralds Changes for Cyber Kill Chain - darkreading.com
In Other News: InjectEave Attack, SIM Swapper Sentenced, Glasswing Findings Review - securityweek.com
Phishing Research Challenges Conventional Security Awareness Testing - securityweek.com
CISA Calls for More Guidance, Less Spin, as Cyber Outages Escalate - darkreading.com
Meta Sued Over Training Data for Its AI and Face-Recognition Systems - wired.com

Cybersecurity Brief – 2025-09-04

Thu, 04 Sep 2025 00:00:00 +0000

Major Incidents or Breaches

Jaguar Land Rover experienced a severe cyberattack, forcing the disconnection of systems and causing significant disruption to both retail and manufacturing operations.
Workiva, a major SaaS provider, disclosed a data breach after attackers accessed customer data via a compromised third-party Salesforce CRM system.
Multiple security firms, including Cloudflare, Palo Alto Networks, and Zscaler, were impacted by a breach involving Salesforce, Salesloft, and Drift, resulting in the exposure of customer contact information and case data.
The Pennsylvania Attorney General’s office confirmed a ransomware attack that caused weeks-long outages of email, phone, and web services. No ransom was paid.
Wiz Research reported a data leak at Chinese AI company DeepSeek, exposing over 1 million sensitive log streams.
The US Department of Justice filed a lawsuit against Apitor Technology for exposing children’s geolocation data to a Chinese third party without consent.
Disney agreed to pay $10 million to settle US FTC claims regarding the improper collection of children’s data on YouTube.

Newly Discovered Vulnerabilities

Google released September 2025 Android updates addressing over 100 vulnerabilities, including at least two zero-days (CVE-2025-2355, CVE-2025-2356) that are actively exploited.
A high-severity use-after-free vulnerability in Google Chrome’s V8 JavaScript engine was patched, preventing potential remote code execution.
CISA flagged an actively exploited vulnerability in TP-Link TL-WA855RE Wi-Fi range extenders, allowing attackers to reset and hijack affected devices. CISA recommends retiring these discontinued devices.
Exploit attempts were observed for CVE-2025-5086, a vulnerability in Dassault DELMIA Apriso used in manufacturing environments.
A WhatsApp bug, in conjunction with an Apple zero-day, is being used in targeted zero-click attacks against iPhone users, potentially for spyware deployment.

Notable Threat Actor Activity

An Iran-linked group, identified as Homeland Justice APT and associated with MOIS, conducted a coordinated spear-phishing campaign targeting over 100 embassy and consulate email accounts globally, aiming to spy on diplomats and international organisations.
Russian APT28 (Fancy Bear) is leveraging new ‘NotDoor’ malware to abuse Microsoft Outlook for covert data exfiltration.
North Korean IT worker scams remain a focus of regional countermeasures, with Japan and South Korea collaborating with private sector partners to disrupt these operations.
Malicious npm packages targeting Ethereum smart contract developers were discovered, using blockchain smart contracts to perform malicious actions on compromised systems.
Automated sextortion spyware, identified as a variant of “infostealer” malware, is taking screenshots and webcam photos of victims viewing adult content for extortion purposes.
Threat actors are using X’s Grok AI assistant to bypass link posting restrictions and distribute malicious links on the platform.
PayPal users are being targeted in a sophisticated email scam with the subject “Set up your account profile.”

Trends, Tools, or Tactics of Interest

Threat actors are rapidly adopting AI-powered offensive tools, specifically HexStrike AI, to exploit newly disclosed n-day vulnerabilities, including Citrix flaws, within days of public disclosure.
Increasing use of generative AI by attackers to craft more convincing phishing emails.
Endpoint security is seeing a shift towards AI-driven solutions, as highlighted in the 2025 Gartner Magic Quadrant, with a focus on countering advanced ransomware and phishing attacks.
Research indicates that compromised routers can remain undetected and accessible on the internet for years, highlighting persistent risks to infrastructure.
Geolocation data is being increasingly weaponised in advanced persistent threat (APT) operations and malware, enabling targeted attacks based on physical location.

Regulatory or Policy Developments

CISA, NSA, and 19 international partners released a joint statement outlining a shared vision for the Software Bill of Materials (SBOM), promoting its role in securing global software supply chains.
The US Department of State announced a $10 million bounty for information on three Russian FSB officers involved in cyberattacks against US critical infrastructure.
Law enforcement, in cooperation with the Alliance for Creativity and Entertainment, disrupted Streameast, the world’s largest illegal sports streaming platform, with arrests made in Egypt.

Cybersecurity Brief – 2025-07-01

Tue, 01 Jul 2025 00:00:00 +0000

Major Incidents or Breaches

Switzerland’s government reported that sensitive data from multiple federal offices was compromised following a ransomware attack on the third-party provider Radix.
Europol, in coordination with Spanish authorities, dismantled a cryptocurrency investment fraud ring responsible for laundering approximately $540 million (€460 million) from over 5,000 victims. Five suspects were arrested.
Over 1,200 Citrix NetScaler ADC and NetScaler Gateway servers remain unpatched against a critical authentication bypass vulnerability, which is reportedly being actively exploited.
The hacker collective Scattered Spider continues its campaign targeting the airline sector, with Microsoft designating the group as a significant cyber threat.
The FBI disclosed that cybercriminals are impersonating health fraud investigators to steal sensitive health information from US citizens.
The US Justice Department revealed that the identities of more than 80 Americans were stolen to facilitate North Korean IT worker scams, with action taken against associated “laptop farms.”
A government report indicated that a hacker was hired by the Sinaloa drug cartel to identify and target individuals connected to the FBI’s investigation, leading to fatal outcomes.

Newly Discovered Vulnerabilities

Vulnerabilities in Airoha Bluetooth chips, used in Sony and Bose earbuds and headphones, could allow attackers to hijack both the audio devices and any connected host systems. Details of the vulnerabilities have not yet been published.
Some Brother printers are affected by a critical security flaw that can be mitigated by changing the default admin password.
Microsoft warned of delays in the delivery of June 2025 Windows security updates due to an incorrect metadata timestamp.

Notable Threat Actor Activity

US agencies (CISA, FBI, NSA, DC3) issued a joint advisory regarding increased Iranian state-sponsored cyber activity targeting US defense, operational technology (OT) networks, and critical infrastructure.
Blind Eagle, a known threat actor, has been linked to the use of the Russian bulletproof hosting service Proton66 for phishing campaigns and remote access trojan (RAT) deployment targeting Colombian banks.
Ongoing AI-themed SEO poisoning attacks are distributing Lumma and Vidar infostealers and other malware via malicious websites optimized for AI-related search terms.

Trends, Tools, or Tactics of Interest

Microsoft Defender for Office 365 now automatically detects and blocks email bombing attacks.
Identity-based attacks, including impersonation and credential theft, continue to rise in frequency and sophistication.
Attackers are increasingly leveraging legitimate system functionality and user credentials, often exploiting “by design” behaviors rather than technical flaws.

Regulatory or Policy Developments Affecting the Security Industry

The Canadian government ordered Hikvision Canada to cease all operations due to national security concerns, though individuals and private businesses may still purchase Hikvision products.
The Berlin Data Protection Commissioner formally requested Google and Apple to remove the DeepSeek AI application from their app stores for GDPR violations.
Microsoft announced it will remove password management functionality from its Authenticator app starting August 1, 2025, as part of a shift towards passkey-based authentication.
Proposed US legislation aims to reduce federal AI funding for states that enact AI regulations, with exemptions for laws addressing unfair practices and child sexual abuse material.

Cybersecurity Brief – 2025-04-23

Wed, 23 Apr 2025 00:00:00 +0000

Major Incidents and Breaches

Marks & Spencer Cyberattack: The UK retailer confirmed a cyberattack affecting operations, including delayed Click and Collect orders. The company implemented temporary operational changes and is responding to the incident.
DeepSeek Breach: An AI-focused organisation suffered a significant breach, with sensitive data appearing on dark web forums, highlighting ongoing risks to AI platforms.
SK Telecom Data Exposure: South Korea’s largest mobile operator disclosed a malware attack that exposed sensitive USIM-related customer data.
Healthcare Ransomware Attacks: Three additional healthcare organisations-DaVita, Bell Ambulance, and Alabama Ophthalmology Associates-were impacted by ransomware, continuing the trend of targeting healthcare.
Baltimore Public Schools Data Breach: A February ransomware attack resulted in the theft of personal information belonging to thousands of students, teachers, and administrators.
City of Abilene Cyberattack: The Texas municipality went offline following a cyberattack, with incident response and investigation ongoing.

Newly Discovered Vulnerabilities

GCP Cloud Composer Privilege Escalation: A now-patched vulnerability in Google Cloud Platform’s Cloud Composer could have allowed attackers to escalate privileges via malicious PyPI packages.
Active! Mail RCE Zero-Day: A remote code execution vulnerability in Active! Mail is being actively exploited in attacks against large Japanese organisations.
Ripple’s xrpl.js Library Compromised: The widely used Ripple cryptocurrency JavaScript library “xrpl.js” was compromised, enabling theft of XRP wallet seeds and private keys.
Cookie-Bite Attack PoC: A proof-of-concept demonstrates how a malicious Chrome extension can steal Azure Entra ID session tokens, bypassing MFA and maintaining access to Microsoft 365 services.
Docker Malware Targeting Teneo Web3 Node: A new malware campaign exploits Docker environments using a previously undocumented technique to mine cryptocurrency by faking heartbeat signals.

Notable Threat Actor Activity

Russian Organisations Targeted by Sophisticated Backdoor: A new backdoor, disguised as secure networking software updates, is being used to target Russian entities.
China-Linked Billbug Group: The Billbug APT (aka Lotus Panda/Blossom, Bronze Elgin) breached multiple organisations in a Southeast Asian country, continuing its targeting of government and telecom sectors.
Russia-Linked Phishing on NGOs: Russian threat actors are conducting phishing campaigns against NGOs connected to Ukraine, using video call lures to compromise Microsoft 365 accounts.
Russian Sabotage Against Dutch Critical Infrastructure: Dutch intelligence reported Kremlin-backed attempts to sabotage critical infrastructure, underlining ongoing state-sponsored cyber risk.
Elusive Comet Hacking Group: This group is exploiting Zoom’s remote control feature in social engineering attacks to steal cryptocurrency from users.

Trends, Tools, and Tactics

Increasing Abuse of AI in Cybercrime: Threat actors are leveraging AI tools to assist in social engineering, phishing, and scam campaigns, as highlighted in Microsoft’s latest Cyber Signals report.
Phishing via Google Sites and DKIM Replay: Sophisticated phishing attacks are abusing Google Sites and DKIM replay to send signed emails that redirect to credential harvesting pages.
Browser Security Concerns: The browser is increasingly the primary endpoint in cloud-native workplaces, yet remains largely unmonitored, presenting a growing attack surface.
Microsoft Secure Future Initiative: In response to the Storm-0558 breach, Microsoft has migrated MSA signing services to Azure confidential VMs and is purging millions of legacy cloud tenants, as part of broader security enhancements.
Android and Chrome Security Updates: Android will introduce auto-reboot after three days of inactivity, and Chrome is dropping third-party cookie prompts while adding IP protection in Incognito mode.
Sensitive Content Protection: Google Messages is rolling out a feature to warn users about potentially explicit images and allow easy blocking, improving user safety.

Regulatory and Policy Developments

Zambia’s New Cyber Laws: Recently signed cyber security and cyber crime acts in Zambia have raised concerns among international observers regarding increased surveillance and potential suppression of dissent.

Other Notable Observations

Edge Vulnerabilities and SMB Threats: Verizon’s report notes a surge in edge-related vulnerabilities, persistent ransomware threats, and ongoing challenges for small and medium businesses (SMBs).
Legal Debates on Device Search Protections: Ongoing discussions about the comparative legal protections of biometrics vs. passcodes for device searches, relevant for privacy and compliance considerations.