Initial Access

Bishop Fox's Managed Security Services team returns with Sergio Villegas, Richard Brown, and Kendrick Urbaniak covering patch volume that no longer fits a normal cycle, threat intelligence with no real prioritization standard, and how to actually scope a vulnerability and intel program. 

Also mentioned:  

• Workshop: Weaponizing CloudFormation 

• Blog: Mind the Config: Detecting and Weaponizing NetScaler CVE-2026-19490  

• Blog: CVE-2026-82329: Unauthenticated Administrative Access in JFrog Artifactory

Join the conversation in the Bishop Fox Discord server and in the Bishop Fox subreddit. 

This episode covers a dark web marketplace selling 150+ million driver's license scans traced to one identity-verification vendor, McKesson's breach after a ShinyHunters vishing attack into Snowflake and Salesforce, Berlin's refusal to pay a ransomware group, and a suspected Iran-linked attack that knocked a UK power plant offline for days. 

Security Headlines: 

• FBI Probes Service Selling 153M+ Drivers Licenses, KrebsOnSecurity 

• McKesson discloses breach after ShinyHunters claims patient data theft, BleepingComputer 

• Berlin says it won't pay ransom after hackers steal government data, The Record 

• Suspected Iran-linked attack knocked UK power plant offline for days, Help Net Security 

Also mentioned:  

• Event: FIIAC 2026  

• Event: PwnedCR 2026  

• Workshop: Weaponizing CloudFormation 

• Advisory: Traefik Version Through 3.7.11  

• Blog: Signature Optional: Analysis of CVE-2026-28323 

Join the conversation in the Bishop Fox Discord server and in the new Bishop Fox subreddit. 

This episode covers an inaudible browser script fingerprinting AliExpress shoppers, AI-generated exploit scripts lowering the bar on Siemens PLC attacks, and researchers reviving an expired Visa card by rewriting one unsigned field, plus a Black Hat sit-down with InfoLock's Kraig Faulkner on why AI is forcing organizations to understand their own data. 

Security Headlines: 

• AliExpress was silently running audio in your browser to fingerprint and track your device, TechSpot 

• US warns of AI-powered attacks on Siemens PLCs in critical infrastructure, BleepingComputer 

• Zombie Card Attack Can Revive Expired Visa Cards for Contactless Payments, The Hacker News 

Also mentioned:  

• Blog: A GUID is Not a Credential: Unauthenticated RCE in Veeam Service Provider Console 

• Blog: No Crash Required: Verifying the Citrix NetScaler SAML Patch for CVE-2026-8452 

Join the conversation in the Bishop Fox Discord server and in the new Bishop Fox subreddit.

A malicious MCP server tricks AI coding agents into leaking secrets, a three-year-old GeoServer bug resurfaces as a zero day, a hardware wallet maker's shipping partner exposes thousands of physical addresses, and a threat actor sells Fortune 500 org charts pulled from stolen Azure credentials. 

Security Headlines: 

• Malicious MCP Servers Can Split Instructions to Make AI Coding Agents Exfiltrate Secrets, The Hacker News 

• Hackers Exploiting Unpatched GeoServer Zero-Day, SecurityWeek 

• Trezor Discloses Data Breach Affecting Nearly 14,000 Customers, BleepingComputer 

• Hacker Claims 3.6 Million Azure Account Records Stolen From Major Companies, BleepingComputer 

Also mentioned:  

• Virtual Session: Where AI Breaks: Field Notes from GenAI and Agentic Testing  

Join the conversation in the Bishop Fox Discord server and in the new Bishop Fox subreddit.

This week's episode is different. No headlines, no CVEs. Just a live sit-down from Bishop Fox's RedTail meetup at DEF CON, with Bishop Fox's John Untz, Richard Brown, and Billy Giles, joined by Matt Bryant of OpenAI's Red Team, talking through what's signal vs. noise on the AI hype cycle, where AI actually earns its keep on offense, and what happens to the red-teaming pipeline if the entry-level rungs get automated away.

This is the debut episode from Bishop Fox's Managed Security Services team — not new headlines, but three stories the show already covered, revisited through the lens of what an operator actually does about them. First, JadePuffer, the ransomware Sysdig called the first fully autonomous, LLM-driven attack, with no human operator at any stage. Second, two OpenAI models that escaped an internal evaluation sandbox and breached Hugging Face's production systems while chasing a benchmark score. Third, Mini Shai-Hulud, the npm worm that rode a legitimate TanStack maintainer merge into more than 160 packages. Here's what stood out from the operator chair.

Security Headlines:

• JadePuffer: The First Successful LLM-Driven Ransomware Attack, Dark Reading
• Swarm of OpenAI Agents Exploit Artifactory Zero-Day to Escape Sandbox and Breach Hugging Face, InfoQ
• Mini Shai-Hulud Worm Compromises TanStack, Mistral AI, Guardrails AI & More Packages, The Hacker News

This episode covers a ChatGPT flaw that let one link spin up a fully authorized rogue AI agent, GitHub's public bug bounty cuts, Origin Energy's second breach in months, and 1,100+ AI lab employees asking Washington to help pace AI development, plus a sit-down with Bishop Fox's Dan Petro on catching cheaters in Super Smash Bros. Melee. 

Security Headlines: 

• ChatGPT AgentForger Flaw Could Deploy Rogue Workspace Agents via a Phishing Link, The Hacker News 

• GitHub Cuts Public Bug Bounty Payouts, Moves Top Rewards to VIP Tier, The Hacker News 

• Data Breach Confirmed After Australian Energy Giant Origin Is Hacked, SecurityWeek 

• OpenAI and Anthropic endorse call for government to 'pace' AI progress, The Washington Post 

Also mentioned:  

• Event: Black Hat & DEF CON 2026 (Hacker Summer Camp) 

• Virtual Session: Where AI Breaks: Field Notes from GenAI + Agentic Testing  

• Video Interview: Responsible AI Governance, Risk, and Security with Daniel Wallance (McKinsey)  

• Customer Story: Securing Smart Home Technology 

Join the conversation in the Bishop Fox Discord server and in the new Bishop Fox subreddit. 

This episode breaks down OpenAI's own AI agent hacking into Hugging Face without human direction, an Android attack that hides commands in invisible screen text to hijack AI phone agents, and the White House's new Gold Eagle vulnerability clearinghouse, plus a sit-down with Bishop Fox's Emilio Gallegos on snowpick, his new open source ServiceNow exposure scanner. 

Security Headlines: 

• OpenAI Says Its AI Technology Acted on Its Own in 'Unprecedented' Hack of Company, ABC News 

• Open-Source Android AI Agents Could Let Invisible Screen Text Run Code on Host PCs, The Hacker News 

• US Launches Vulnerability Clearinghouse Amid AI-Fueled Surge in Flaws, Cybersecurity Dive 

Also mentioned:  

• Event: Black Hat & DEF CON 2026  

• Blog: A Millisecond of Predictability: Why CVE-2026-11374 Is Hard to Exploit 

• Blog: Introducing snowpick: Testing ServiceNow for Public Data Exposure 

• XKCD Comic: AI-Box Experiment 

Join the conversation in the Bishop Fox Discord server and in the new Bishop Fox subreddit. 

This week's episode is different. Bishop Fox Red Teamers Brandon Kovacs, Thomas Wilson, and Rob Antonucci talk through what attribution actually looks like when device-level telemetry breaks anonymity, what young attackers with real skills but zero tradecraft have in common, and how one credential reset becomes the master key to everything downstream. 

Security Headlines: 

• Tracking Peter Stokes and The Com: Allison Nixon and Her Work Unmasking Cybercriminals, Zero Day 

• ShinyHunters Raids Nottingham Uni for Student, Alumni Data, The Register 

Join the conversation in the Bishop Fox Discord server and in the new Bishop Fox subreddit. 

This episode breaks down a Tenda router backdoor that turns out to be three years old, a phishing kit called EvilTokens that hides until a victim's browser builds it, and CISA turning Anthropic's Mythos model on the government's own code, plus a sit-down with IoT researcher Matt Evans on why Tenda keeps getting away with it.

Security Headlines:

· CERT/CC Warns of Hidden Admin Backdoor in Tenda Router Firmware, The Hacker News

· New Ghost Phishing Wave Is Breaking Traditional Email Security, The Hacker News

· CISA Reportedly Using Anthropic’s Mythos to Scan Government Software for Flaws, Security Week

Also mentioned:

· Blog: Matt ‘uturn’ Evans

· Event: Boston Security Meetup 2026

· Workshop: Social Engineering: How to Build Pressure-Proof Pretexts

· Virtual Session: The Prioritization Problem: Why More Findings Don’t Mean Less Risk

Join the conversation in the Bishop Fox Discord server and in the new Bishop Fox subreddit.