Sign in

Mike Sec

@sudoinit0.bsky.social
73 followers 211 following 177 posts

Nat Sec | Info Sec | CTI Nerd

PostsRepliesMedia
Mike Sec @sudoinit0.bsky.social · 08/10/2026
Interesting from a cyber threat intelligence perspective, but likely common these days and not worthy of “headlines”.
000
Mike Sec @sudoinit0.bsky.social · 07/10/2026
Not gonna lie - this may be one of the more novel, simple, but effective C2 channels I’ve seen since actors began hiding comms in Britney Spears photos decades ago. cyberscoop.com/poellm-malwa...
cyberscoop.com
PoeLLM malware has assembled a sweeping botnet, taking technical cues from a poem
More than 3,400 servers have been compromised by malware that hides its infrastructure coordinates in a poem.
020
Mike Sec @sudoinit0.bsky.social · 02/10/2026
Citrix Security Awareness Month.
010
Reposted by Mike Sec
GreyNoise @greynoise.io · 29/09/2026
A timeline of Citrix NetScaler CVE-2026-88771, from the CVE reservation on Sep 10 to public disclosure on Sep 27, including the exploitation attempts GreyNoise observed on Sep 24. 🔗 Full analysis: www.greynoise.io/blog/swarmin...
094
Mike Sec @sudoinit0.bsky.social · 29/09/2026
Citrix continues to get beat up worse than the Eagles… labs.watchtowr.com/here-we-go-a...
labs.watchtowr.com
Here We Go Again (Citrix NetScaler DTLS Preauth Memory Overflow CVE-2026-88772)
Part 1 of this week's saga can be found here. This research is a glimpse into the capabilities that power our Preemptive Exposure Management solution, enabling organizations to rapidly react to emerg...
001
Mike Sec @sudoinit0.bsky.social · 28/09/2026
100%.
000
Mike Sec @sudoinit0.bsky.social · 28/09/2026
Always entertaining…albeit for sad reasons… labs.watchtowr.com/oh-look-the-...
labs.watchtowr.com
Oh Look, The Foot Gun Went Off Again (Citrix NetScaler PreAuth Command Injection CVE-2026-88771)
God damn it, we're back in the room again. We'll probably write more here later, but for now, deal with this picture of our favorite software dev, who works at Citrix (we imagine). Citrix, before yo...
000
Mike Sec @sudoinit0.bsky.social · 25/09/2026
Some poignant thoughts by HF on the impact of their public disclosure. x.com/clementdelan...
x.com
clem 🤗 (@ClementDelangue) on X
What we learned from being the first company to disclose an agent cyberattack
000
Mike Sec @sudoinit0.bsky.social · 25/09/2026
Possible Kiteworks targeting underway or soon: www.heise.de/en/news/Immi...
heise.de
Imminent Zero-Day Attack: KiteWorks Urges Customers to Shut Down Servers
Manufacturer writes to its customers that they have concrete indications from law enforcement about an attack. Large companies are also affected in this country.
000
Mike Sec @sudoinit0.bsky.social · 22/09/2026
The more interesting takeaway from this post is really how the LLM quorum decides what steps to take next. There is no single indicator that fully identifies the architecture, only behaviors.
000
Mike Sec @sudoinit0.bsky.social · 22/09/2026
AI enabled malware introduces a new aspect that needs to be considered: effort displacement - transferring an entire phase of the attack from the operator to the system. blog.talosintelligence.com/the-closed-q...
blog.talosintelligence.com
The Closed Quorum: Inside the first reported autonomous AI C2 implant
CLOSEDQUORUM, a malware binary discovered through Cisco Talos’ CAIRN project, exhibits fully autonomous command and control (C2). It represents a shift in effort displacement for attackers, in which e...
000
Mike Sec @sudoinit0.bsky.social · 21/09/2026
The U.S. government has multiple programs across the FBI, CISA, NSA, and elsewhere working with industry, so that’s not net new. If anything, and it can’t be underscored enough, the need to rebuild government capacity after the hemorrhaging over the last two years should be priority #1.
000
Mike Sec @sudoinit0.bsky.social · 21/09/2026
One year spent on this study and zero tangible conclusions. Governments have struggled for years with how to impose costs on Chinese operations, and no answer is in sight (also, saying you should counter cyber with cyber tells me you have no experience in this space). www.cfr.org/articles/the...
cfr.org
The U.S. Is Highly Vulnerable to Cyber Threats From China. Here’s What It Should Do | Council on Foreign Relations
Protecting vulnerable infrastructure will require a multipronged approach to make the United States a tougher target.
110
Mike Sec @sudoinit0.bsky.social · 15/09/2026
Another example of how AI is a lot like junior analysts.
010
Reposted by Mike Sec
brian nam-sonenstein @bson.bsky.social · 08/09/2026
dear god.. a flavor scientist at the Cheetos factory has published a blog post saying no one is prepared for how "dangerously cheesy" they are
12134392
Reposted by Mike Sec
Mark Kelly @mkyo.bsky.social · 09/09/2026
🚨🌙 Today we are releasing a new @threatinsight blog on BlueMoon, an exploit kit that chained “patch-gap” Chrome zero-days with a Windows zero-day and had indicators of AI-assisted development. It was used by at least four espionage-motivated threat actors since late August.
175
Mike Sec @sudoinit0.bsky.social · 08/09/2026
Interesting research on the vulnerabilities llms.txt files introduce and their related findings, primarily how “content on your website is no longer content — it’s operational input.” medium.com/@alonhertz1/...
medium.com
Data Became Code: We Ran Code Inside Fortune 500s Using Files They Published for AI Agents
It Was That Easy
000
Mike Sec @sudoinit0.bsky.social · 02/09/2026
Marketing dressed in security clothes. openai.com/index/path-t...
openai.com
Path to Astra: critical capabilities and frontier safeguards
Astra is the first OpenAI model to meet the Critical cybersecurity capability threshold under the Preparedness Framework, with stronger safeguards for release.
010
Mike Sec @sudoinit0.bsky.social · 26/08/2026
Neckties were once used to help keep necks warm. Over time they lost the functional use and became decoration men still have to wear for whatever reason. I feel that’s where we’re at with virtualization from a security aspect. Once functional, now decoration. blog.trailofbits.com/2026/08/26/v...
blog.trailofbits.com
VMs won't contain cyber-capable agents
You can no longer assume a mere VM will contain a sufficiently advanced AI agent.
000
Mike Sec @sudoinit0.bsky.social · 19/08/2026
I…I don’t think that’s how it works…
000
Mike Sec @sudoinit0.bsky.social · 13/08/2026
In other words, Section 3, “Implementation Guidance”, will be the tar pit from which this will likely never leave.
000
Mike Sec @sudoinit0.bsky.social · 13/08/2026
Given the legal hurdles, inter-organization and interagency logistics required, and operational deconfliction that will need to take place prior to execution, I find it unlikely that this will get off the ground prior to the end of the administration, if ever.
010
Mike Sec @sudoinit0.bsky.social · 12/08/2026
The two layer probabilistic decision engine is both impressive and worrying. Many have noted that AI enabled attacks are/will be “loud”; however, this provides evidence for how actions taken may be far more deliberate and tougher to detect moving forward. dreamgroup.com/blog/inside-...
dreamgroup.com
Inside a Multi-Agent AI Framework Used to Compromise Government Entities in Asia | | Dream Security Blog
000
Mike Sec @sudoinit0.bsky.social · 11/08/2026
While I agree that “vulnpocalypse” was hyperbolic, on the flip side of that I would argue that, outside of speed and tangential connections, we as defenders have no consistently objective way of identifying AI enabled attacks (which most are these days), from non-AI attacks to apply metrics.
000
Mike Sec @sudoinit0.bsky.social · 10/08/2026
Kimsuky (or threat actors writ large) using local models to likely avoid detection is not surprising; however, how they’re plugging stolen documents into a RAG solution for question answering and likely attack development is interesting. www.genians.co.kr/en/blog/thre...
genians.co.kr
Kimsuky Integrates AI into Attack Operations, From AI-Generated Decoy Documents to a Local LLM
Operation GitPower, a Kimsuky campaign, uses AI-generated lure documents and GitHub/GitLab-based C2 infrastructure to deliver malicious payloads.
000
Mike Sec @sudoinit0.bsky.social · 08/08/2026
Now, you’re not going to believe this, but “Each compromised substation associated with a wind farm contained a FortiGate device functioning as both a VPN concentrator and a firewall.” cert.pl/uploads/docs...
cert.pl
100
Mike Sec @sudoinit0.bsky.social · 04/08/2026
Bro, this is my gear - for real. blog.talosintelligence.com/keep-going-b...
blog.talosintelligence.com
“Keep going, bro. You’ve got this!” A data-driven look at how adversaries are weaponizing AI
Talos has collected prompt logs from threat actor endpoints running various applications, such as Claude Code, CodeX, Cursor, or Gemini. This blog is an analysis of the ways we've seen bad actors leve...
000
Reposted by Mike Sec
Microsoft Threat Intelligence @threatintel.microsoft.com · 04/08/2026
Microsoft Threat Intelligence is tracking active Mini Shai-Hulud npm supply chain attacks in which a threat actor compromised trusted maintainer accounts to distribute credential-stealing malware.
Supply chain
163
Mike Sec @sudoinit0.bsky.social · 04/08/2026
The Administration’s approach to AI is akin to the Catholic church’s stance during the Renaissance - do everything you can to block progress, to include making it illegal, because you don’t know how to handle change that may contradict previous world order. www.nytimes.com/2026/08/04/t...
nytimes.com
White House Whipsaws Silicon Valley (and Itself) Over A.I. Rules
The Trump administration has struggled over how to approach to “open source” models, which are freely available to download and favored by Chinese companies.
010
Mike Sec @sudoinit0.bsky.social · 25/07/2026
While an advancement, it likely wasn’t worth the hyperbolic reaction most had when Kimi K3 was released. www.nist.gov/news-events/...
nist.gov
UK AISI / CAISI Preliminary Assessment of Kimi K3's Cyber Capabilities
The UK Artificial Intelligence Security Institute (UK AISI) and the U.S.
000
Mike Sec @sudoinit0.bsky.social · 24/07/2026
Time to update those actor mappings… cloud.google.com/blog/topics/...
cloud.google.com
Updated Cyber Threat Actor Naming System | Google Cloud Blog
000
Mike Sec @sudoinit0.bsky.social · 14/07/2026
I’ve read this multiple times and still don’t understand what they’re trying to say or do… www.whitehouse.gov/releases/202...
whitehouse.gov
White House Launches Gold Eagle Initiative for Unprecedented Cybersecurity Vulnerability Coordination
President Trump’s bold vision to secure and accelerate American artificial intelligence (AI) innovation is being actioned through the creation of “GOLD
010
Mike Sec @sudoinit0.bsky.social · 13/07/2026
Long before Salt Typhoon the Russians were targeting routers leveraging similar, if not identical, TTPs - and they continue to do so today. media.defense.gov/2026/Jul/09/...
media.defense.gov
000
Mike Sec @sudoinit0.bsky.social · 25/06/2026
“The AI bubble is bursting because organizations bought the wrong story. They thought they were buying replacement labor. What they actually bought was an expensive, overconfident junior assistant” 🎯 www.thecipherbrief.com/the-ai-bubbl...
thecipherbrief.com
The AI Bubble and the Growing National Security Problem
Rising AI costs may lead to the real supply-chain risk, a silent shift toward Chinese models.
000
Reposted by Mike Sec
Joe Slowik @pylos.co · 24/06/2026
Cultural exchange
0344
Mike Sec @sudoinit0.bsky.social · 18/06/2026
“The shift is not a contraction of the role. It's a redistribution within it…The role isn't disappearing. It is getting harder.” Great blog on the reshaping of the SOC analyst role in the age of AI. binarydefense.com/resources/bl...
binarydefense.com
The Analyst's New Job Description
The deterministic work is moving to agents, which is good, because that's where most of the burnout was coming from. The judgment work is staying with…
000
Reposted by Mike Sec
Catalin Cimpanu @campuscodi.risky.biz · 18/06/2026
WTF is this? Did Lab52 get hacked by a Chinese APT? lab52.io/blog/from-ch... archive.ph/XTK3j web.archive.org/web/20260618...
063
Mike Sec @sudoinit0.bsky.social · 17/06/2026
Friendly reminder that, while AI has been a tool for attackers to commoditize and scale operations, basic OPSEC still matters. research.openanalysis.net/claude/codex...
research.openanalysis.net
Captured Logs Reveal Hackers Using Claude and Codex to Breach Companies
Full agent sessions captured on a compromised host turned honeypot offer an unprecedented look at how attackers are using AI in real-world intrusions.
000
Mike Sec @sudoinit0.bsky.social · 16/06/2026
I think they need to correct the title: “The AI race won’t be won by the best model, but by the fastest military - and the DoD keeps shooting themselves in the foot.” www.thecipherbrief.com/the-ai-race-...
thecipherbrief.com
The AI Race Won't Be Won by the Best Model—But by the Fastest Military
Washington has identified artificial intelligence as a defining national security challenge. But as China rapidly integrates AI into military operations and state strategy, experts warn America's grea...
000
Mike Sec @sudoinit0.bsky.social · 11/06/2026
The timeline with associated reports, the impact that US politics likely had on group coverage, and the marked shift in TTPs are all notable. APT28 leveraging an LLM for operational logic in LameHug likely didn’t get enough attention last year. blog.sekoia.io/apt28-an-evo...
blog.sekoia.io
APT28, an evolution of tradecraft
Context Sekoia’s Threat Detection & Research (TDR) team has been tracking APT28 for several years. The intrusion set, also known as Fancy Bear, Forest Blizzard, Sofacy, Pawn Storm or Sednit and public...
000
Mike Sec @sudoinit0.bsky.social · 10/06/2026
I feel this should be a lesson to all the publications that quickly yelled “fire” without evidence or nuance in their language, but it won’t..
000
Mike Sec @sudoinit0.bsky.social · 09/06/2026
Is there a story or rumor I’m missing that their gender even matters versus just their position…?
000
Mike Sec @sudoinit0.bsky.social · 06/06/2026
Anyone that has been to a Preakness knows that those are prime for races across the tops. My heart goes out to those inside when someone falls through.
000
Mike Sec @sudoinit0.bsky.social · 06/06/2026
APT 10 was a fairly pervasive campaign across a number of MSPs and CSPs. It’s reasonable to believe that if they didn’t look for them back then, they’re still in there as their activity is likely baked into any baselining that has been done since. www.bankinfosecurity.com/ex-threat-in...
000
Reposted by Mike Sec
Volexity @volexity.com · 04/06/2026
@volexity.com has published details from an incident response engagement in September 2025 involving multiple #BRICKSTORM variants deployed by a threat actor that Volexity tracks as VerdantBamboo. [1/4]
volexity.com
VerdantBamboo: Just Another BRICKSTORM in the Firewall
In September 2025, Volexity conducted an incident response engagement that began after suspicious network traffic was observed from a Linux-based virtual machine appliance on a customer’s network. The...
186
Mike Sec @sudoinit0.bsky.social · 04/06/2026
Not too often MI5 co-seals an advisory, but always good to see. www.mi5.gov.uk/five-eyes-jo...
mi5.gov.uk
Five Eyes Joint Bulletin - Safeguarding Our Secrets | MI5 - The Security Service
Members of the Five Eyes intelligence partnership release a joint bulletin warning of the threat posed by China's military intelligence services on Western professional networking sites and online job...
000
Mike Sec @sudoinit0.bsky.social · 31/05/2026
While this is almost certainly happening more than is being reported or identified, it’s interesting to see how they were able to piece together the artifacts to make the conclusion that this was an autonomous attack. www.sysdig.com/blog/ai-agen...
sysdig.com
AI agent at the wheel: How an attacker used LLMs to move from a CVE to an internal database in 4 pivots | Sysdig
A Sysdig TRT investigation reveals their first seen LLM-agent-driven intrusion: from CVE-2026-39987 to internal database exfiltration in under one hour.
000
Mike Sec @sudoinit0.bsky.social · 26/05/2026
If you don’t store logs, you can’t find malicious activity. If you’re not finding malicious activity, you have good cybersecurity. cyberscoop.com/white-house-...
static.klipy.com
Think About It Meme with Reece Simpson
ALT: Think About It Meme with Reece Simpson
100
Mike Sec @sudoinit0.bsky.social · 25/05/2026
Been saying this for a while - cyber is like artillery, it’s a supporting arm. Battles may be decided by it, but won’t win them by itself.
000