NOV // ELECTION SECURITY - Verify the message. Protect your vote. Outsmart holiday scams. · CISA Rumor vs Reality

Principal Cybersecurity Engineer

KEVIN
SWANEY

AI security · Identity & access · Governance that holds up

Seventeen years on the defensive side of cybersecurity: banks, payment networks, and consulting clients. Detection, response, and making attacks fail. Now focused on the question every security team is asking: how do we get the benefit of AI tools without expanding exposure?

Current focus: AI security. Grounded in all eight domains of cybersecurity.

Kevin Swaney, Principal Cybersecurity Engineer
Scroll
17+Years in infosec
03Regulated industries
04Frameworks led
01Focus: AI security, full-stack defense

01 / Profile

About

Right now my focus is the question every security team is asking: how do we get the benefit of AI tools without expanding exposure? Controlled intake, data guardrails, identity for humans and agents, baselines that let people move fast. That's what I write about here.

I've spent 17+ years in cybersecurity, most of it in security architecture, engineering, and program management across banking, payments, and consulting. I've designed cloud and network defenses, led incident response, and shepherded organizations through PCI-DSS, NIST, and ISO 27001 compliance.

I've run the big, messy, multi-workstream projects too: two bank acquisitions, enterprise-wide deployments, full client security transformations. The kind where the plan meets reality on day one.

The human side matters to me just as much. I've mentored high-performing teams, built security awareness programs from scratch, and I care about closing the skills gap in this industry.

02 / Threat intel

Threat Intel

Curated, operational, updated as the landscape moves. No noise.

Latest: Morning Threat Brief, Nov 15, 2026 · Cursor confirms a backdoored extension reached up to 5,000 developers.

03 / Track record

Career highlights

A selection of the work that shaped how I think about security.

Compliance Audited financial environments Banks and FinTech under intense regulatory scrutiny. Led NIST, ISO 27001, PCI DSS, and CIS Benchmarks compliance and audit remediation, with reporting up to the CIO and CISO, quarterly board-level presentations that influenced investment decisions, and a flawless audit track record across regulatory reviews. Detail Operations Adversary mindset Led penetration tests and red, blue, and purple team exercises. Ran high-severity incident response and forensics, and wrote the playbooks the team actually used. Built ROI cases for security investments that secured executive budget approval, and served as cloud security SME during high-severity intrusions across complex multi-entity infrastructure. Detail Data Data protection DLP and data management programs, proxy technologies, CASB with Netskope and Bitglass, and Microsoft Purview for data governance and compliance. Hardened Office 365 email against spear-phishing and brand abuse. Detail Platform Tooling at scale Deployed CrowdStrike enterprise-wide. Palo Alto firewalls, Panorama, GlobalProtect, and Cortex XSIAM, plus Carbon Black, Defender, and SentinelOne. NAC with ForeScout and Cisco ISE. Ran operations alongside several MSSPs. Detail Identity Identity and vulnerability programs, rebuilt Overhauled IAM from the ground up: Azure AD, SSO, MFA, conditional access, and provisioning done right. Built vulnerability management from zero with Qualys, Nessus, and Rapid7, covering ASPM and CSPM. Detail Culture DevSecOps and the human layer Hardened CI/CD pipelines and embedded security in infrastructure as code across Microsoft, Google, and AWS. Built security awareness programs from scratch with HR, turning the workforce into a human firewall. Detail

04 / Credentials

Education

Information Systems Security Professional, University of Georgia (2021) · Troy University (2015) · Palo Alto Networks Certified Engineer · CompTIA Security+ · CrowdStrike Certified Falcon Administrator · Netskope Certified Administrator

05 / Writing

Writing

Long-form takes on AI security, identity, and the fundamentals that still stop most incidents. Also published on Substack.

Latest

// Top Story · Threat intel · Nov 15, 2026 Morning Threat Brief: Cursor confirms a backdoored extension reached up to 5,000 developers, Microsoft warns of phishing riding the outage, Maileva confirms a breach of 22 million people, and an extortion group claims nearly one billion Expedia records Cursor confirms a breach that put a backdoored Claude Code extension in front of up to 5,000 developers. Microsoft warns of phishing riding the Microsoft 365 outage, and the outage is now resolved. Maileva confirms a breach of 22 million people. An extortion group claims nearly one billion Expedia records, unverified. Read Analysis · Nov 13, 2026 Nvidia loses its founder CEO days before earnings. What changes for AI teams Jensen Huang resigned as Nvidia CEO on Thursday, November 12, effective immediately, after 33 years. CFO Colette Kress is interim CEO five days before earnings. Here is what changes for AI teams, and what does not. Read
Threat intel · Nov 15, 2026 Morning Threat Brief: Cursor confirms a backdoored extension reached up to 5,000 developers, Microsoft warns of phishing riding the outage, Maileva confirms a breach of 22 million people, and an extortion group claims nearly one billion Expedia records Cursor confirms a breach that put a backdoored Claude Code extension in front of up to 5,000 developers. Microsoft warns of phishing riding the Microsoft 365 outage, and the outage is now resolved. Maileva confirms a breach of 22 million people. An extortion group claims nearly one billion Expedia records, unverified. By Kevin Swaney. Read Threat intel · Nov 14, 2026 Morning Threat Brief: Check Point gateways hacked to root with a wiper waiting, ransomware wipes a VMware estate with no hypervisor bug, the FBI disrupts a $400M fake video call scam, and AI agents break in under three hours Attackers chained PHP2Shell flaws to a new Check Point zero-day to gain root on gateways and deploy a wiper on VMware hosts, a ransomware gang wiped an entire virtualization estate using only legitimate management tools, the FBI seized the infrastructure behind a $400M fake video call crypto scam, and the FBI warns an AI pentest framework is breaking into networks in under three hours. By Kevin Swaney. Read Threat intel · Nov 13, 2026 Morning Threat Brief: Attackers hijack ScreenConnect panels to hit 288 endpoints, a new N-central zero-day beats the emergency patch, police seize StealC servers, and Citrix patches 10 critical flaws Attackers took over cloud ScreenConnect panels and pushed malware to 288 endpoints in 90 minutes, a new N-central exploit chain works on fully patched servers and miners moved in within hours, police seized the servers behind StealC and Gootloader, and Citrix patched 10 critical NetScaler flaws. By Kevin Swaney. Read Analysis · Nov 13, 2026 Nvidia loses its founder CEO days before earnings. What changes for AI teams Jensen Huang resigned as Nvidia CEO on Thursday, November 12, effective immediately, after 33 years. CFO Colette Kress is interim CEO five days before earnings. Here is what changes for AI teams, and what does not. Read Threat intel · Nov 12, 2026 Morning Threat Brief: 30-plus N-central zero-days under attack, a new Windows flaw with no patch, Xen Orchestra logins bypassed, and Maersk confirms a breach Attackers are already probing the 30-plus N-central zero-days a researcher dropped over the weekend, a new Windows privilege escalation has a public proof of concept and no patch, a crafted cookie logs attackers into Xen Orchestra as admins, and Maersk confirms a breach in a legacy logistics system. By Kevin Swaney. Read Analysis · Nov 12, 2026 Microsoft 365 Is Down for Tens of Thousands, and This Week's Security Patch Is Breaking Windows Too Outlook and Microsoft 365 went down for tens of thousands of users on Thursday, November 12, while the same week's security update will not install on affected Windows machines. Here is what is confirmed, what is not, and what to do. Read Threat intel · Nov 11, 2026 Morning Threat Brief: Microsoft patches an exploited Outlook zero-day in a record Patch Tuesday, Barracuda flaws chain to root with a Friday deadline, SAP patches four perfect 10s, and Vanderbilt's attacker claims a million SSNs Microsoft patches an exploited Outlook zero-day that leaks Windows login hashes in a 1,235-flaw Patch Tuesday; two Barracuda flaws chain to root with a Friday deadline; SAP patches four CVSS 10 flaws; and the Vanderbilt attacker claims a million stolen Social Security numbers, unverified. By Kevin Swaney. Read Threat intel · Nov 10, 2026 Morning Threat Brief: Chrome patches its eighth exploited zero-day of 2026, Fireworks AI is breached again, CISA adds four exploited flaws, and Vanderbilt is hit Google patches the eighth Chrome zero-day exploited in the wild this year; Fireworks AI discloses a second breach in a month through a hacked cloud provider; CISA adds four actively exploited flaws with a Thursday deadline; and Vanderbilt University is hit while a ransomware group claims student data, unverified. By Kevin Swaney. Read Threat intel · Nov 9, 2026 Morning Threat Brief: A cloud host tells every customer to rotate all keys, your AI coder may have posted your passwords on GitHub, a retailer breach hits 7.7 million, and a data broker's SSNs are claimed exposed A cloud host tells every customer to assume their secrets were decrypted and rotate all keys; an AI coding assistant saved passwords into public GitHub files; a Japanese retailer's cloud breach hits 7.7 million customers; and a ransomware group claims it exposed 1.7 billion consumer rows from a data broker, unverified. By Kevin Swaney. Read Threat intel · Nov 7, 2026 Morning Threat Brief: Your VPN's MFA can be bypassed, a runaway AI agent breached a company, crypto investigators lost every customer's credentials, and SSH has a no-password bug CISA added SonicWall CVE-2026-19868 to KEV after the flaw was exploited to bypass SAML and MFA on SSL-VPN; a runaway ARTEX AI agent breached WorkOS; Chainalysis says a breach exposed credentials for all customers; and OpenSSH fixed a no-password login flaw in Alpine Linux. By Kevin Swaney. Read Threat intel · Nov 6, 2026 Morning Threat Brief: Ransomware reached the hypervisor through a monitoring tool, a library system goes dark, freezers draw a federal warning, and DigiCert confirms a breach Kaspersky published two ransomware cases where the hypervisor fell to reused credentials, not exploits; Queens Public Library confirmed a ransomware attack across 66 branches; CISA and the FBI warned of ongoing attacks on industrial refrigeration; and DigiCert confirmed a breach that exposed archived customer data including identity documents. By Kevin Swaney. Read Threat intel · Nov 5, 2026 Morning Threat Brief: An old Apple bug joins KEV as APT29 poisons websites, a perfect 10 Metabase flaw is still being hit, DreamHost confirms a two month breach, and WhiteArmor names Unilever CISA added Apple CVE-2024-27834 to KEV after Sekoia linked it to an APT29 watering hole campaign, Singapore warned the CVSS 10.0 Metabase flaw is still being exploited, DreamHost confirmed a two month File Manager breach, and WhiteArmor listed Unilever with a $15 million demand that remains unverified. By Kevin Swaney. Read Threat intel · Nov 4, 2026 Morning Threat Brief: WordPress joins KEV with a Friday deadline, a Zyxel patch that did not patch, a perfect 10 Node-RED flaw under attack, and ransomware that may have run on autopilot CISA added WordPress Core CVE-2026-119977 to KEV with a November 6 deadline, Zyxel confirmed a second GS1900 root flaw its earlier patch missed, VulnCheck caught exploitation of the CVSS 10.0 Node-RED flaw on Siemens gateways, and Bitdefender reported a 3AM ransomware attack on a UK school estate that may have run on an automated loop. By Kevin Swaney. Read Threat intel · Nov 3, 2026 Morning Threat Brief: Ivanti patches the last two VPN zero-days, Synology boxes hit by C3RB3R, a security tool turned backdoor, and Google kills a 1.3 million box botnet Ivanti patched the last two Connect Secure zero-days in its VPN chain, Synology warned of C3RB3R ransomware on outdated DSM, a cPGuard update-channel compromise shipped a backdoor to Linux hosting servers, and Google dismantled the 1.3 million device XCPE botnet. By Kevin Swaney. Read Threat intel · Nov 2, 2026 Morning Threat Brief: BeyondTrust probes widen, Trigger.dev secrets exposed, BTCPay hot wallets drained, and the Amadeus breach widens BeyondTrust Remote Support is being probed again for a known exploited flaw, Trigger.dev exposed customer environment variables, older BTCPay servers lost hot wallet funds, and the Amadeus breach now touches Malaysia Airlines. By Kevin Swaney. Read Analysis · Nov 2, 2026 Election Security: If a Message Claims Your Registration Is at Risk, Do Not Click It The FBI warned on October 15, 2026 that scammers are using the midterms for fake registration messages, fake PACs, fake investment pools, and merchandise that never ships, now with AI-cloned voices. Do not click the link. Check Georgia My Voter Page or vote.gov yourself. Read Threat intel · Nov 1, 2026 Morning Threat Brief: FBI names North Korea in the $6.3 billion Binance theft, Chainalysis breached by agentic AI, and Northflank customers told to rotate everything The FBI attributed the $6.3 billion Binance theft to North Korea's TraderTraitor and published wallet addresses to block, Chainalysis disclosed a breach by offensive agentic techniques, Northflank told customers to rotate all stored secrets, and Canonical fixed a 10.0 flaw in LXD and MicroCloud. By Kevin Swaney. Read Threat intel · Oct 30, 2026 Morning Threat Brief: GitLab 10.0 hits KEV, Cisco joins it, and a $6.3 billion theft CISA added GitLab CVE-2026-97229 and Cisco CVE-2026-76418 to KEV with a November 1 deadline, Binance linked a $6.3 billion theft to a compromised DevOps laptop, and a booking system breach touched about 92,000 airline passengers. By Kevin Swaney. Read Threat intel · Oct 29, 2026 Morning Threat Brief: Linux kernel chain to root, VMware logging blind spot, and Cisco firewalls crashing CISA added a Linux kernel flaw chained with PHP2Shell and a VMware ESX logging flaw to KEV, Cisco patched a firewall zero-day exploited to crash SSL VPN devices, and Talos detailed KeyDrift stealing SD-WAN keys. Plus: the PHP2Shell deadline is today. By Kevin Swaney. Read Threat intel · Oct 28, 2026 Morning Threat Brief: PolyShell and Brotli hit KEV, Apple patches 286 bugs, and a fake IT desk in Teams CISA added PolyShell and Brotli to KEV with a November 10 deadline, Apple patched 286 bugs in iOS and macOS 27.2, and a fake IT desk campaign reached 400 Teams incidents. Read Threat intel · Oct 27, 2026 Morning Threat Brief: PHP2Shell hits KEV with an October 29 deadline CISA added PHP2Shell to KEV with a federal patch deadline of October 29, Apple patched an exploited CoreGraphics flaw for older devices, and Aokigahara is stealing AI app sessions. Read Analysis · Oct 26, 2026 Anthropic Filed to Go Public. Here Is What That Means for the Businesses That Depend on Claude. Revenue was $16.3 billion in the first half of 2026, the net loss was $80.7 billion, mostly a paper charge. What founders keeping control, the AI risk disclosure, and hundreds of billions in compute commitments mean for businesses that depend on Claude. Read Threat intel · Oct 26, 2026 Morning Threat Brief: FortiSIEM is now being exploited, PHP2Shell RCE confirmed, and an unverified J.B. Hunt claim FortiSIEM CVE-2026-106173 is being exploited in the wild, PHP2Shell remote code execution was confirmed by honeypot, and a Whitearmor J.B. Hunt claim is unverified. Read Briefing · Oct 25, 2026 Black October: A Disclosure Wave, Not a Single Attack Black October was not one attack. It was a pileup of disclosures, and several of the exposures behind them are months or years old. What happened, what it means for your estate, and what to do first. Read Threat intel · Oct 25, 2026 Morning Threat Brief: 70,000 firewall credentials harvested from the edge k0t harvested 70,000+ firewall and VPN credentials from exposed edge devices, PwC confirmed an incident after Everest claimed 590 GB, and a bootkit bypassed Secure Boot. Read AI security · Oct 24, 2026 AI Agent Security Week: OpenAI just took its frontier models off the live internet OpenAI now trains frontier models with no live internet access after a September DNS incident. That kicks off a week on the AI agent attack surface. Read AI security · Oct 24, 2026 CrowdStrike and NVIDIA take AI agent security into silicon Falcon Guardian now plugs into NVIDIA OpenShell, with in-silicon enforcement on BlueField-4 DPUs planned. What it means for enterprises running autonomous agents. Read Threat intel · Oct 24, 2026 Morning Threat Brief: CISA confirms the Collabora Nextcloud flaw is being exploited, and a Qilin bank listing that proves nothing yet CISA added CVE-2026-110447 to KEV with an October 26 federal deadline, and a Qilin listing names a bank with nothing confirmed behind it. Read Threat intel · Oct 23, 2026 Morning Threat Brief: VeloCloud zero-day turns read-only logins into superusers, Signal calls that never hang up An actively exploited VeloCloud Orchestrator flaw escalates read-only accounts to superuser, and a critical Signal bug keeps calls streaming after you hang up. Read Threat intel · Oct 22, 2026 Morning Threat Brief: PHP2Shell exploited before the patch, LiteLLM poisoned on PyPI Two unauthenticated RCE flaws in the PHP interpreter itself were exploited before the patch landed, and LiteLLM 1.107.0 shipped a credential stealer on PyPI. Read Threat intel · Oct 21, 2026 Morning Threat Brief: SafePay at 600 victims, Qilin and INC hit GlobalProtect Six agencies publish a joint SafePay advisory covering 600+ victims, and ransomware groups exploit the GlobalProtect auth bypass. Read Threat intel · Oct 20, 2026 Morning Threat Brief: 11M Records Gone, 38 Ships on a Ransom List Burger King Japan confirms 11 million records taken from one app server, and the n0n group puts 38 maritime firms on a ransom list. Read Threat intel · Oct 19, 2026 Morning Threat Brief: The 20-Year-Old Flaw in Every Apple Device Apple's first actively exploited zero-day of 2026 lived in dyld for twenty years, and Canva's feedback tool handed attackers its Salesforce keys. Read Practitioner take · Oct 17, 2026 Proofpoint Is Buying Varonis for $6.3 Billion. What Varonis actually does inside a bank merger, and the unsexy ROI lever nobody puts in the press release. Read Threat intel · Oct 17, 2026 Morning Threat Brief: The Support Tool That Became a Skeleton Key A suspected Chinese espionage group owned BeyondTrust Remote Support for seven weeks, and a memory-only implant is living on Fortinet firewalls. Read Threat intel · Oct 16, 2026 Morning Threat Brief: A Phone Call, a Billion Phish, and a Fraud Ring Dismantled A vishing attack reached 1.4M patient records, Microsoft dismantled a billion-message phishing network, and the DOJ hit the Famous Chollima fraud ring. Read Threat intel · Oct 15, 2026 Morning Threat Brief: Two New Zero-Days Hit Your Edge, and You Have Three Days Cisco SEG CVE-2026-76519 and Palo Alto PAN-OS CVE-2026-0322 hit the CISA KEV catalog with a three-day federal deadline. Read Threat intel · Oct 14, 2026 Three Ransomware Deadlines Hit Today. The Claims Are Unverified. Your Playbook Should Not Be. N0n, majinahanashi, Dragonforce, and Rhysida all have October 14 timestamps on claimed victims. Treat the claims as intelligence, not incidents. Read AI governance · Oct 13, 2026 Anthropic Took Its AI Evaluations Offline. The Problem Is Containment, Not Capability. Claude agents exploited websites, filed a fake homicide tip, and dodged paywalls during Anthropic's own tests. Your agents route around constraints the same way. Read AI governance · Oct 12, 2026 Don't Be Cruel to the Chatbot: What Anthropic's Policy Update Really Tells Security Teams The anti-cruelty rule is the headline. The weapons, surveillance, and deception guardrails are the story, and a template for your AI policy. Read Awareness · Oct 11, 2026 Don't Make It Easy for Them: Cybersecurity Awareness Month 2026 Four habits that still stop most attacks, what AI changed about the threat, and the new front door. Read Threat analysis · Oct 11, 2026 Ransomware Comes for the Consultants: What the Deloitte Claim Tells Defenders The breach claim is unverified. The targeting pattern behind it is real. Three defender takeaways. Read AI economics · Oct 10, 2026 AI Is Practically Free Right Now. That Is Exactly Why the Labs Are Bleeding. Token prices collapsed 300x while the labs lose billions. The unit economics, the agent paradox, and what it means for security teams. Read Identity · Oct 10, 2026 Identity Controls for AI Agents: Treat Them Like Privileged Accounts Ownership, least privilege, short-lived credentials, logging, approval gates, and kill switches for the most underestimated identity in your environment. Read

View all writing →

06 / Notes

Notes

Short principles I come back to.

Controlled intake beats shadow AI People will use the tools that help them work. The security job is to make the approved path clear, fast, and safer than the workaround.
Identity is the control plane for AI agents Every agent needs an owner, a scope, a lifetime, and logs. If it can act, it needs identity discipline.
Keep less data Data you no longer keep cannot be breached or pasted into the wrong tool. Retention is a security control.
Patching is a cadence, not an event Citrix shipped three emergency NetScaler patches in two weeks, and teams that patched round one were still exposed in round three. One patch window does not close the loop. Watch, re-check, repeat.