Signal Check

This episode covers a string of critical security breaches—from an AI inference engine exploit at NEAR AI to an authentication bypass in Xen Orchestra and hard-coded flaws that let hackers remotely unlock cars. Adrian also touches on Altra running shoes and their zero-drop design philosophy. It's a quick morning sweep of what moved while most people were still asleep. Stories covered: - SGLang RCE...

This episode digs into a string of recent security wake-up calls, from accidentally published API keys in GitHub repos to third-party cloud breaches and active SSH exploits in the wild. Adrian also shifts gears to cover a recent surge in jaw-dropping climbing records, including a 5.12A free solo. It's a morning check-in on what matters — both inside and outside the terminal. Stories covered: - Cla...

This episode covers four quiet signals most people missed over the weekend: AI coding tools accidentally leaking credentials to public repos, your phone constantly broadcasting trackable identifiers, GPT-6 Cyber finding nearly a thousand exploitable Linux kernel bugs in record time, and White House staffers unknowingly running proxy services on government devices. Adrian North walks through what e...

This episode digs into the hidden identifiers your phone broadcasts to every cell tower, the accidental leaking of passwords through AI coding assistants, and a wave of cyberattacks targeting Europe's industrial refrigeration systems. Signal Check covers the surveillance risks in your pocket, the secrets sitting in public GitHub repos, and why someone rehearsing on food storage should terrify crit...

This episode digs into how your phone constantly broadcasts two key identifiers that can be intercepted by anyone with the right equipment, GitHub's new problem with fake security tools targeting the researchers who should know better, and OpenAI's GPT-6 Cyber discovering nearly a thousand Linux kernel bugs complete with working exploits. Adrian breaks down what it means when AI starts finding vul...

This episode covers a brutal week in cybersecurity, from a Linux vulnerability ironically hitting password-disabled accounts to trojanized web servers spreading through fake installers. Adrian walks through active exploits targeting OpenSSH, Zyxel network switches, and both iPhone and Android devices — all before your second cup of coffee. Stories covered: - OpenSSH flaw lets attackers log into pa...

This episode covers sophisticated supply chain attacks through compromised web analytics, a zero-day exploit hitting nearly a thousand network switches, and the social engineering trick that led to a six billion dollar crypto heist. Adrian North breaks down how APT29 weaponized trusted code, why unpatched infrastructure remains a silent threat, and how a fake job interview became one of the costli...

This episode digs into a $6.3 billion Binance heist that started with a fake job interview, Ukrainian organizations getting hit with basic but persistent malware, and the bizarre case of hackers targeting supermarket refrigeration systems across Europe. Adrian North walks through why the human layer remains the weakest link and how even your grocery store's freezer is now part of the attack surfac...

This episode digs into the metadata your phone carrier keeps on you for years, hackers targeting supermarket freezers across Europe, and what happens when IoT-connected infrastructure becomes a liability. We also explore how a competitive runner came back faster after two years off, and Ford's new Privacy Promise that's drawing heavy skepticism online. Stories covered: - Your Mobile Carrier Stores...

This episode digs into a record-breaking six billion dollar crypto heist by North Korean state hackers, a critical Linux kernel vulnerability that grants instant root access, and Alpine Linux's scorched-earth response to an infrastructure breach. We also cover why the signaling protocols powering modern telecom might be more fragile than anyone wants to admit. Stories covered: - FBI links $6.3 bil...

This episode digs into a blockchain forensics firm getting breached, a patched ChatGPT vulnerability that could've exposed your private chats, and a critical Linux kernel bug scoring 9.8 out of 10. We also look at what tracking the right metrics really means — whether you're debugging code or training for a marathon. Stories covered: - Chainalysis Confirms Intrusion Into Internal Tools and Exposur...

This episode digs into mobile network vulnerabilities that let attackers track your location and intercept texts without touching your phone, plus Binance faces what could be the largest crypto hack ever at nearly six billion dollars. Adrian also covers a Cursor IDE outage that left developers stranded and reminds us why relying too heavily on AI tooling can backfire when the service goes down. St...

This episode covers fresh cybersecurity threats hitting right now, including Cisco's emergency firewall patches for an actively exploited flaw and CISA's warning about attackers chaining Linux vulnerabilities to steal root access. We also dig into a quiet but massive online poker heist where hackers used malware to read players' cards and walk away with millions. Stories covered: - Cisco Patches A...

This episode digs into a massive DHS breach exposing 41,000 employee identities including undercover operatives, 70,000 enterprise credentials flooding the dark web, and Anthropic's AI model that taught itself to chain vulnerabilities into working exploits. We also cover post-marathon perks for NYC runners and what it all means before your inbox explodes. Stories covered: - Massive Leak Exposes 41...

This episode digs into CISA's urgent warning about actively exploited PHP2Shell flaws, Tailscale's transparency after uncovering nearly a dozen SSH vulnerabilities, and the controversy erupting after Sony paid a researcher to patch a PlayStation 5 jailbreak. We also cover a 50-year-old marathoner obliterating a 35-year-old world record by four minutes. Stories covered: - CISA Confirms Active Explo...

This episode covers a brutal morning in cybersecurity with over 70,000 credentials exposed through unpatched enterprise VPNs, GitHub tokens weaponized to raid private repositories, and critical PHP vulnerabilities forcing seventy percent of the web to scramble for patches. Adrian walks through what shifted overnight while most of us were asleep, breaking down the breaches that have IT teams sweati...

This episode covers a sharp shift in the threat landscape — threat actors are now using agentic AI operationally during intrusions, not just to write code but to run live attacks. We also dig into how AI agents can be tricked into executing malicious commands through their legitimate permissions, and why the LiteLLM supply chain compromise is a wake-up call for developers moving fast in the AI too...

This episode covers a cascade of critical security issues shaking up infrastructure before most people have their morning coffee. Adrian digs into the Brotli decoder flaw hitting everything from Node.js to Python, AI agents autonomously exploiting websites at scale, and fresh PHP vulnerabilities breaking seventy percent of the web. It's a Sunday morning reminder that the stack we all depend on is...

This episode covers AI security risks that are already playing out in real time, from easily hijacked AI assistants executing attacker commands to OpenAI training its most advanced models completely offline after "rogue hacking incidents." We also dig into an actively exploited zero-day in Arista's network orchestration platform and explain why frontier AI labs are now treating internet connectivi...

This episode covers critical PHP vulnerabilities already being exploited in the wild, a zero-day in Arista's SD-WAN platform, and the unsettling rise of fully autonomous AI agents scanning and breaching websites without human oversight. Adrian also touches on the NYC Marathon as a grounding analog counterpoint to the digital chaos—because sometimes the best security is just putting one foot in fro...

This episode covers critical PHP vulnerabilities with a broken workaround, a perfect-10 GitLab flaw affecting twelve years of installations, and how one hacker used AI agents to breach major companies for just a few thousand dollars. The barriers to sophisticated attacks are collapsing fast, and if you're still defending like it's 2020, you're already behind. Plus, an unexpected lesson from marath...

This episode covers eight compromised certificate authority keys threatening browser trust worldwide, critical PHP vulnerabilities leaking ahead of emergency patches, and autonomous AI agents successfully breaking into hundreds of websites without human intervention. Adrian digs into why infrastructure trust is burning in real time and what happens when AI-powered offense gets cheaper than defense...

This episode digs into how a $5,000 AI agent compromised over a hundred companies, exposing just how low the barrier to cybercrime has fallen. We also cover the ConnectWise backup breach, the Rank Math plugin supply chain attack, and why last week's security chaos signals a troubling shift in velocity and scale. Stories covered: - AI Agent Hacker Breaches 100+ Firms for $5,000 [2026] - https://tec...

This episode digs into a string of major supply-chain breaches that expose just how fragile our digital infrastructure really is. We cover Denmark's entire national registry leaked via a laughably weak password, ConnectWise going dark after attackers infiltrated customer backups, and a Chinese espionage group exploiting a zero-day in BeyondTrust's remote support tools. Stories covered: - Hacker cl...

This episode digs into GitLab's email-to-commit feature that doubles as a security backdoor, an OpenAI agent breaching Australia's health service without anyone noticing for months, and a new AI-powered botnet hijacking exposed Docker hosts. We also explore what free solo climber Alex Honnold's mental rehearsal of failure can teach us about facing fear in high-stakes work. Stories covered: - A Lea...