In the latest Threat Source, Hazel channels her inner Spielberg to explore why humans are delightfully irrational, reminding us that while security best practices are simple in theory, they’re a lot harder to pull off when you’re busy deali

A new approach to reverse engineering, developed by Cisco Talos, integrates local artificial intelligence agents with traditional analysis tools. This method utilizes the VB6 disassembler, vbdec, to expose parsed data through a live Component Object Model (COM) interface. This allows security analysts to use natural language prompts to automate complex tasks such as decompiling functions and constructing call graphs, transforming the disassembler into an interactive data server.
This methodology offers significant advantages for security professionals. It empowers analysts to create custom workflows dynamically, eliminating the need to wait for new features from software vendors. Furthermore, it addresses privacy concerns by keeping sensitive binaries on the analyst's local machine, as both the AI agent and the disassembler operate on the same workstation. This architectural shift demonstrates that any analysis tool with structured data accessible via a graphical user interface can be enhanced for automated tasks, potentially saving defenders considerable time.
The developers of analysis tools are encouraged to expose their application data through external scripting interfaces like COM or other inter-process communication protocols. For those analyzing VB6 binaries, enabling remote scripting in vbdec and directing a local AI agent towards the provided operator briefing can automate various tasks. Security teams are advised to embrace this paradigm shift, allowing AI agents to handle repetitive and extensive work while human analysts focus on higher-level analysis.
In other cybersecurity news, the group ShinyHunters has claimed responsibility for a hack on the Council of Europe, threatening to release over 297GB of stolen data. Separately, a widespread credential-harvesting operation has compromised more than 30,000 Fortinet devices globally, targeting firewalls and VPN gateways.
A fileless malware known as Phantom Stealer has been observed targeting browser credentials. This malware employs anti-analysis techniques, including in-memory execution, to evade detection. In a different incident, a security researcher discovered a flaw in an internal FIFA World Cup system that allowed unauthorized individuals to control and view the TV stream of all games.
The FBI has established a dedicated facility called the Kinetic Cyber Range, a replica small town designed to simulate real-world cyber attacks. This facility, operational since February 2025, includes various structures and infrastructure mimicking a typical U.S. community.
Cisco Talos also highlighted several resources for cybersecurity professionals. A podcast episode, "Patching in the dark," discusses the complexities of managing unknown threats and the practical challenges of patching in intricate environments. Another resource details Cisco Talos Threat Hunting's use of hypothesis-driven methods and multi-domain telemetry correlation to detect sophisticated threats. Additionally, a discussion with Tony Giandomenico, Senior Director of Product Management with Cisco Talos, explores endurance in cybersecurity and the discipline required for success. Talos will also be present at upcoming events such as Black Hat USA and DEF CON 34.
The most prevalent malware files identified through Talos telemetry in the past week include samples detected as Win.Worm.Coinminer, W32.C0AD494457-95.SBX.TG, and Win.Tool.Procpatcher.
A weakness has been identified in Tenda CP3 27.5.57.101. This issue affects some unknown processing of the file Net/NetCheckPing.cpp. This manipulation of the argument interface_name/host causes os command injection. The attack can be initiated remotely.
A security flaw has been discovered in Tenda CP3 27.5.57.101. This vulnerability affects the function SystemAsh of the file Apis/system.c of the component Kylin. The manipulation of the argument AlarmVoiceURL results in os command injection. It is possible to launch the attack remotely.

OpenAI has announced a $1 billion commitment to provide subsidized access to its Daybreak AI cybersecurity tools for under-resourced critical infrastructure defenders. The initiative, named Daybreak for Frontline Defenders, will offer AI models, training, and technical support over the next six months, prioritizing water and wastewater utilities, electric grid operators, and local government entities. This move aims to equip organizations with limited budgets and staff against increasingly sophisticated cyber threats.

Attackers are exploiting a new unpatched vulnerability in Magento Open Source and Adobe Commerce that lets them run malicious code on an online store's server without logging in, Dutch e-commerce security company Sansec said in an advisory published on September 5. Sansec, which discovered the flaw and named it StyleSmuggler, said attacks started on September 4. "Sansec is publishing early
In BPF instructions that load/store a value from/to a scratch memory register the register index is an unsigned 32-bit integer and must not exceed 15, but libpcap BPF interpreter does not validate the value. In particular uncommon use cases a crafted filter program can cause the interpreter to try reading and writing the OS process memory in the 16GiB starting at the current stack frame on 64-bit architectures and in the entire address space on 32-bit architectures.

Attackers are exploiting two new PaperCut flaws to steal credentials and gain privileged access in education-sector attacks across the U.S. and Europe. Attackers are exploiting two recelty disclosed PaperCut flaws, CVE-2026-81578 and CVE-2026-82078, in attacks targeting schools and other education organizations in the U.S. and Europe, as reported by TheHackerNews. Arctic Wolf researchers observed