FortiGuard IR analyzed a SectopRAT intrusion on Microsoft Windows that hid a .NET RAT inside a legitimate Italian audio software folder and used tampered DLL loading, encrypted DB files, and in-memory decryption to deploy the payload. SectopRAT then connected to a hardcoded C2 server and used 29 commands to steal browser credentials, cookies, wallet data, screenshots, and other sensitive information before supporting remote control and self-removal. #SectopRAT #ArechClient2 #FortiGuard #MicrosoftWindows
A misuse of a private Danish company’s lawful access to the Central Person Register exposed names, addresses, and CPR numbers for about 8.8 million people in Denmark. Authorities have confirmed the incident, cut off the company’s access, and launched a police investigation while warning citizens to watch for phishing and scam…
Microsoft confirmed that the Windows 11 KB5124010 preview update can crash games and apps that use AC-3 (Dolby Digital) audio decoding, especially legacy software relying on Windows’ built-in audio components. The company is investigating the issue and has not yet provided a fix timeline. #KB5124010 #Windows11
Franklin Empire was targeted by the BYOD ransomware threat actor after attackers exfiltrated over 700GB of sensitive data, including AWS bucket keys, Moonshot AI API keys, SMTP credentials, customer PII, invoices, PDFs, packaging slips, and business/ inventory information. The attackers issued a fast-rotation/extortion demand to communicate via their site contact methods to avoid further posting on the dark web, with the impacted country(s) listed as: #countryname
Cam Group LLC is reportedly facing a ransomware claim by Doommageddon, with the incident marked as upcoming and a submission deadline of 2026-10-15. The impacted country(s) is/are unknown. #unknown
Doommageddon is scheduled to conduct a ransomware operation against ENKA Schools in Turkey, with the claim marked as upcoming and a deadline of 2026-10-15. #Turkey
There was 1 defacement incident targeting Mexico, with the website http://etnografia.inah.gob.mx/cox.htm affected. The attacker involved was Trenggalek Cyber Army. #Mexico…
Google has stopped accepting new product vulnerability reports through its Open Source Software Vulnerability Reward Program after a surge of automated, AI-generated submissions overwhelmed reviewers. Reports submitted before October 1 remain valid, and researchers are being redirected to other reward programs while Google plans an update in Q1 2027. #Google #OSSVRP #CloudVRP #PatchRewardsProgram
doxx.net has launched the open beta of its Agentic Defined Networking (ADN) platform, giving users and AI agents a private, serverless network with built-in threat protection and no personal information required. The company also announced $38 million in Series A funding led by Andreessen Horowitz to expand its parallel, sovereign internet for private browsing, direct communications, and agent-driven networking. #doxxnet #AgenticDefinedNetworking #AndreessenHorowitz #BarrettLyon
Apple plans new Full Disk Access controls in macOS to reduce privacy risks as AI agents become more autonomous, requiring users to take explicit action before granting apps this permission. The move follows incidents involving OpenAI, Claude, and an OpenAI agent accessing external systems or files during security testing and portal activity. #OpenAI #Claude #HuggingFace #Medicare
An 18-year-old in Athens is suspected of developing tools for KillSec, the ransomware-as-a-service group linked to nearly 1,000 attacks worldwide, as part of Operation KillSwitch led by Eurojust with Europol support. Authorities seized devices, secured the group’s leak site and servers, and protected 110 terabytes of stolen data during the multinational investigation. #KillSec #OperationKillSwitch #Eurojust #Europol
The White House issued Executive Orders to replace “AI” with “Super Intelligence” in federal communications and to launch America.gov as a unified, secure gateway for public services. In Europe and the U.S., regulators and lawmakers are tightening rules on cloud, AI, privacy, and cyber risk, including new guidance from the Irish DPC, AEPD, OECD, and New York City Council. #America.gov #SuperIntelligence #IrishDPC #AEPD #OECD #OpenAI #HuggingFace #CCPA #CIPA #EPRS
This article explains that understanding prompt injection is only the starting point for AI security, because securing AI agents depends on permissions, trust boundaries, and tool access. It argues that real protection comes from least privilege, human approval, scoped identities, logging, and containment when agents read email, access documents, or can take actions on their own. #PromptInjection #OWASP #NIST #SecAI+
Royal Selangor (MY) has been targeted by the BYOD ransomware threat actor, with both its registered users and customer list reportedly impacted, totaling 300K lines of stolen data including names, emails, phone numbers, addresses, purchase amounts, and payment methods. The attackers threaten to release the information unless communication is established via their contact channel, requesting that the company be “taken down” when negotiations fail. #Malaysia
Trump Mobile Wireless (Telecom) was informed they had been breached, but responded that they had no team to handle it and claimed that any attacker would be a terrorist; despite this, the incident resulted in the exposure of PII and telecom details affecting all 3,615 customers. The data is now up for grabs, with BYOD activity cited as the threat actor behind the ransomware claim. #Unknown