SEC595: Applied Data Science and AI/Machine Learning for Cybersecurity Professionals


Experience SANS training through course previews.
Learn MoreLet us help.
Contact usBecome a member for instant access to our free resources.
Sign UpWe're here to help.
Contact UsOn Wednesday, March 11, 2026, Michigan-based medical technology supplier Stryker reported a "global network disruption to [its] Microsoft environment as the result of a cyberattack." According to news sources, the attack wiped devices managed by Stryker's network, including both corporate and personal computers and mobile devices, and led to thousands of employees being sent home and cut off from corporate email systems and other work platforms. A source known to Brian Krebs believes that the affected devices were managed and remotely wiped by Microsoft Intune, a cloud-based endpoint management service. Stryker believes the incident is contained, and has found no evidence of ransomware or malware. Ongoing updates to the security notice state that business continuity measures are in place; the electronic ordering system is undergoing restoration and orders will be processed once communications return; email and phone communication with Stryker employees is safe; and there is no impact to Mako, Vocera, or LIFEPAK devices, nor to the LIFENET service. Users should contact their electronic patient care reporting (ePCR) vendor or hospital administrator if their ePCR or hospital systems have temporarily paused transmissions. A social media post on the same day alleges responsibility for the attack by a threat group called Handala, linked with the Iranian Ministry of Intelligence and Security; some Stryker employees working at the company's hub in Cork, Ireland, have reported a Handala logo appearing on affected devices. External advisors and cybersecurity experts are aiding ongoing investigation into the scope, nature, and operational and financial impacts of the attack, according to the company's 8-K filing, but Stryker foresees ongoing disruptions and limitations, with an unknown timeline for full restoration.
It might be time for MDM/MAM product managers (MS Intune/JAMF/NinjaOne/etc) to ask themselves if they want to perhaps maybe think about including DEFAULT "Do not nuke the entire fleet" rate limits and approval workflows, since the remote wipe feature is not really meant for fleet-wide wiping, but rather selective activity related to lost/stolen/decommissioned devices. There is no reason for this default behavior of allowing fleet-wide nukeware enablement via APIs, other than to serve bad actors. Please talk about this with your MDM vendors. An identity impersonation attack should not result in the ability to wipe an entire fleet, just like breaking into your bank account should not equate to a full removal of all deposited funds to another bank, without being bothered and re-authenticated out of band by the original bank. While it is possible to add multi-admin approval to Intune, for example, this should be the default for anything beyond a certain number of endpoints to be wiped, with alarms going off for attempts at large-scale device wiping as an additional default. It does appear that there are some rate limits related to Intune wiping via APIs, but they are easily bypassed, as Stryker learned, with some parallelized batching, etc. See Intune Multi-admin approval here: https://learn.microsoft.com/en-us/intune/intune-service/fundamentals/multi-admin-approval

MDM/EMM is not just for phones anymore; it can manage laptops, desktops and more. It's a powerful tool for your team. Let's go with the scenario that your device management software is compromised. This is a case where you not only must have strong MFA, but also must limit access to the management interfaces as well as ensuring you have sufficient logging. Verify your forensic capabilities, including making sure you have at least six months of logs available.
An example of an evildoer using what’s available to them (i.e., living off the land), causing mischief and mayhem. Unified Endpoint Management (UEM) products like Intune ease the security and management journey for IT and security staff. What isn’t stated in the article is that the tool isn’t available to the average user. It probably goes without saying that the loss of IT and/or security staff admin credentials enabled the attack. That’s a different problem for the company.

A prime example that demonstrates how a conflict in one region can have far reaching cybersecurity consequences to organisations that are located away from that region. Your risk management processes should include monitoring the geopolitical landscape and reviewing your risk and threat profile based on how geopolitics will impact your organisation. Many reports highlight that those behind this attack used Stryker's own Intune environment to remotely wipe devices. During a conflict, destruction and disruption will be the primary motivators of attackers, so plan your business continuity and cyber resilience accordingly. Finally, should you wish to enhance the security around changes to your Intune environment, you can configure Access policies to require multi-administrator authorisation.

A historical reminder: In March 2010 the Stuxnet malware targeted Iranian nuclear power plants and spilled over to globally impact some OT systems using targeted controllers. The current war in Iran means precaution levels should be assessed and raised where appropriate.
Stryker
SEC
Dark Reading
Krebs on Security
HIPAA Journal
The Record
Nextgov/FCW
The Register
BleepingComputer
Apple has released updates for older versions of iOS and iPadOS to address several vulnerabilities leveraged by the Coruna exploit kit; the fixed vulnerabilities include a use-after-free issue in the kernel (CVE-2023-41974); a type confusion issue (CVE-2024-23222) in WEbKit; a use-after-free issue in WebKit (CVE-2023-43000); and an unspecified vulnerability (CVE-2023-43010) in WebKit, the description for which notes that processing maliciously crafted web content may lead to memory corruption and that the issue was addressed with improved memory handling. In a write-up of Coruna earlier this month, Google Threat Intelligence Group states that the exploit kit "contain[s] five full iOS exploit chains and a total of 23 exploits." Coruna has 12 associated CVEs, three of which (CVE-2021-30952, CVE-2023-43000, and CVE-2023-41974) were added to the US Cybersecurity and Infrastructure Security Agency's (CISA's) Known Exploited Vulnerabilities (KEV) catalog on March 5, 2026. (See NewsBites Vol. 28, No. 18 from Tuesday, March 10, 2026, for more details.)

This update goes all the way back to the iPhone 6s, supporting devices released more than 10 years ago. The Coruna exploit kit began exploiting these vulnerabilities as early as 2024, and since then, Apple has released updates to patch them. Right now, the exploits have only been used in targeted attacks. But it is reasonable to expect these exploits to “trickle down” to less sophisticated actors and be used more widely as miscreants better understand the underlying vulnerabilities.

While it’s great that Apple backported the fixes to iOS 15 and 16, you really need to be on one of their current supported versions. These versions are technically unsupported. iOS 26 works on the iPhone 11 + and iPads from 2019 on; that’s over six years. I know some folks are hanging onto their iPhone 6s, 7s, & 8s (2015, 2016 & 2017). It’s time.

This story has many legs, covered for about a week in various places. I highly recommend reading through the Coruna items. However, outside of that story, this patch is important for all Apple Products. One of the rarer cases I recommend patching your devices for is a potential for exploitation. It’s good to see older devices getting patched, given the seriousness of the campaign.
One of the very few times that AAPL has updated end-of-life versions of their software products. This speaks volumes to the severity of the vulnerabilities and ability to compromise. If you’re still running older versions, take advantage of the free update; or, better yet, update to a new product.
The Hacker News
BleepingComputer
SecurityWeek
Apple
Apple
NIST
On Wednesday, March 11, 2026, Europol, in collaboration with law enforcement agencies from multiple countries as well as Eurojust, executed Operation Lightning, which disrupted operations of the SocksEscort proxy network. Law enforcement agents took down and seized 34 domains and 23 servers in seven countries. US law enforcement agents froze US $3.5 million in cryptocurrency. The investigation began in June 2025 in response to a Europol Joint Cyberaction Task Force (J-CAT) case that revealed a botnet comprising numerous infected devices, many of which were home and small business routers. SocksEscort sold access to this botnet to cybercriminals wishing to hide their actual IP addresses and locations. Countries participating in the operation include Austria, Bulgaria, France, Germany, Hungary, the Netherlands, Romania, and the US. The operation was also supported by private sector partners, including Lumen’s Black Lotus Labs and the Shadowserver Foundation.

The malicious proxy service compromised consumer routers, modems, and IoT devices. While this service is now gone, you need to make sure that your devices are not compromised and that you're good from the next attempt. The fix is to make sure that you're not only keeping the firmware updated, but also check that you're running supported devices. Where you've got ISP provided gear, find out what their practice is for EoL devices. You may have to proactively request updated hardware or buy your own.

These types of residential proxies are extremely important to attackers to hide and evade detection. The problem lies in the security of many residential systems. Until we have hardened, and I mean extremely hardened devices, this will continue to be a problem. It’s just too simple for these edge devices to be compromised and then used to hide attacker traffic.
Congratulations to the men and women in blue for the successful takedown. Unfortunately the reality is that the proxy service lasted for five years and it took investigators about a year to shutter the operation. Given that home users typically hang on to SoHo routers and other IoT devices well past their expiration date, the evildoers will reform for another five year run.

Well done to Europol and all those involved in this operation. While arresting and detaining those behind these schemes is the best solution, disrupting their operations and seizing their assets is an effective measure and raises the costs for the criminals.

The DHS alert last year pointed to attackers exploiting unpatched vulnerabilities in ConnectWise and Fortinet software. There have been a lot more patches to edge and security software since then — use this item to justify prioritizing patching.
Europol
CyberScoop
The Register
BleepingComputer
Justice
A February 2025 cyberattack that targeted Milwaukee-based Bell Ambulance compromised protected health information (PHI) belonging to more than 235,000 individuals. Bell Ambulance detected anomalous activity on its network on February 13, 2025. At that time, a third-party investigation revealed that PHI belonging to 114,000 individuals had been compromised. Those people were notified in April 2025. A year-long internal review of the incident, which concluded on February 20, 2026, revealed that the breach affected PHI of 237,830 individuals; letters have been sent to all those affected. The compromised data include first and last names, birth dates, Social Security numbers, driver’s license numbers, financial account information, medical information, and health insurance information.

This attack was claimed by the Medusa ransomware gang, which targets critical infrastructure organizations, healthcare, and government institutions in Minnesota, Illinois, and Texas, and even NASCAR. I am glad the event was investigated thoroughly, and affected individuals were notified shortly after discovery, but we either need to get faster at this, or every one of us needs to assume our data has been compromised and protect our credit and identity regardless.
Three conclusions to draw from the success ransomware gangs have had over the past 3-4 years: 1) Almost any sized organization can fall victim; 2) Companies don’t do a particularly good job of managing the PII/PHI data they retain; and 3) Companies are somewhat ambivalent in getting data breach notifications out to victims. This is but the latest example.

Turns out that healthcare data is important to attackers and can be monitored.
On Tuesday, March 10, Microsoft released updates to address more than 80 vulnerabilities in its products. For the first time in six months, none of the vulnerabilities fixed in Microsoft's monthly security release are being actively exploited. Two of the vulnerabilities, an SQL Server Elevation of Privilege Vulnerability (CVE-2026-21262) and a .NET Denial of Service Vulnerability (CVE-2026-26127), were publicly known prior to the release. Eight of the vulnerabilities are rated critical. Of those, five affect cloud tools and do not require customer action to patch; the other three critical flaws affect Excel and Office.

Knock on wood, this month's fixes don't include any zero-day flaws; February had five. Make sure you are tracking the nine fixes for Edge, as well as the separate patch for Windows Server 2022 released last week. Don't lose track of the updates from Adobe, which include updates to Acrobat and Adobe Commerce.
SANS ISC
Krebs on Security
CyberScoop
The Hacker News
The Register
SecurityWeek
Earlier this week, SAP, Adobe, HPE, Cisco, and others released security updates. SAP released 15 new security notes, including two that address critical vulnerabilities: a code injection vulnerability in SAP Quotation Management Insurance application (FS-QUO, CVE-2019-17571) and an insecure deserialization in SAP NetWeaver Enterprise Portal Administration (CVE-2026-27685). Adobe released updates to address 80 vulnerabilities, including four critical flaws in Adobe Commerce and Magento Open Source and five critical flaws in Adobe Illustrator. HPE released fixes for five vulnerabilities in Aruba Networking AOS-CX, including a critical authentication bypass issue in Web Interface (CVE-2026-23813). Cisco published four advisories to address multiple vulnerabilities in Cisco IOS XR Egress Packet Network Interface Aligner Interrupt, Cisco IOS XR Software Multi-Instance Intermediate System-to-Intermediate System, Cisco IOS XR Software, and Multiple Cisco Contact Center Products.

The Cisco XR vulnerabilities have CVSS scores of 7.4 to 8.8. Since these are core/edge devices, don't pass go, update them. Don't overlook your Contact Center products, these have XSS and SSRF flaws, which aren't currently being exploited. Better safe than sorry after someone reverse engineers the update.
The Hacker News
SAP
Onapsis
Adobe
HPE
Cisco
Heise
On Wednesday, March 11, the US Cybersecurity and Infrastructure Security Agency (CISA) added a critical remote code execution via expression injection vulnerability in n8n to the Known Exploited Vulnerabilities (KEV) catalog. The flaw could be exploited to execute arbitrary code with privileges of the n8n process. The vulnerability was first disclosed in December 2025; CISA has confirmed that it is now being actively exploited. US Federal Civilian Executive Branch (FCEB) agencies have until March 25 to mitigate the issue. On Monday, March 9, CISA added three vulnerabilities to KEV: one high-severity flaw in Ivanti Endpoint Manager (CVE-2026-1603) and one high severity server-side request forgery issue in Omnissa Workspace One UEM, formerly known as VMware Workspace One UEM (CVE-2021-22054) — both of which have mitigation due dates of March 23 — and one critical deserialization of untrusted data vulnerability in SolarWinds Web Help Desk (CVE-2025-26399) that has a mitigation deadline of Thursday, March 12.

CVE-2025-68613, n8n privilege escalation flaw, has a CVSS score of 10.0, and of the 203,000 n8n users, over 103,000 appear still vulnerable. Although you need to be a user to exploit the flaw, there are other n8n issues which warrant updating now. The easy fix is to move to v1.122.0. Even with the lapse in federal funding, it's good to see CISA is keeping the KEV updated, despite a banner warning to the contrary. Keep watching it.
The Register
The Hacker News
BleepingComputer
GitHub
NIST
NIST
NIST
NIST
The CERT Coordination Center (CERT/CC) has published a notice describing a new attack technique that uses a malformed ZIP file to bypass antivirus detection. Metadata indicates which processes must be performed on files by the antivirus software before a security scan (such as decompression), so if an attacker falsifies the compression method flag, the software will scan the file as though it contains uncompressed data when in fact the contents are still DEFLATE compressed, causing the scan to miss any signatures. When the cyclic redundancy field (CRC) is also mismatched, common tools such as 7-Zip, unzip, or WinRAR will be unable to extract the contents, but a "purpose-built loader that ignores the declared method" can still decompress the malformed file and extract a potentially malicious payload. Chris Aziz of Bombadil Systems discovered this technique and dubbed it “Zombie ZIP.” Aziz states that all but one of 51 tested engines on VirusTotal can be bypassed using this technique. CVE-2026-0866 has been reserved by a CVE Numbering Authority (CNA), and CERT/CC tracks this as VU#976247, but there is disagreement about the categorization — Cisco has issued a statement that while they will take it into consideration for future releases, "this is not considered a vulnerability, but rather, a hardening suggestion." CERT/CC recommends that scans not rely on declared metadata, and also that antivirus and EDR tools include "more aggressive detection modes to validate compression method fields against actual content characteristics, and flag inconsistencies for further inspection."

This is not a “real” vulnerability. Reading the malicious archives would require custom software. If an attacker is able to install custom software to unpack malware, they may as well use encryption.

Due to the need for a custom decoder, the risk from this type of zip file is nominal. Even so, it's not a bad idea to remind people about caution with attachments, particularly with ones which seem "broken." SANS ISC handler Didier Stevens published tools that can analyze the malformed ZIP files if you're researching them. https://isc.sans.edu/diary/Analyzing+Zombie+Zip+Files+CVE20260866/32786
CERT/CC
GitHub
SANS ISC
SC Media
BleepingComputer
Researchers at Lumen's Black Lotus Labs have observed the infrastructure of a botnet that comprises routers and other network devices. In all, more than 14,000 devices have been infected with malware the researchers call KadNap. More than 60 percent of the infected devices are located in the US. The Black Lotus Labs team writes, "KadNap employs a custom version of the Kademlia Distributed Hash Table (DHT) protocol, which is used to conceal the IP address of their infrastructure within a peer-to-peer system to evade traditional network monitoring." The infected edge devices use that protocol to find and connect with command-and-control (C2) servers, which are fairly well camouflaged from defenders.

The KadNap botnet was sold through Doppelganger, and Lumen has blocked access to their C2 servers. We still have work to do. The majority of the compromised devices are Asus routers. If you've got one, make sure it's not harboring new features, set a strong password, apply firmware updates, and disable WAN access to the management interface. Replace them before they go EoL. Have your enterprise team incorporate the IoCs from Lumen to make sure that you don't have any matches.
Lumen
Ars Technica
The Hacker News
SANS Internet Storm Center StormCast Friday, March 13, 2026
IoT Device Discovery; Apple Patches; Veeam Patches
https://isc.sans.edu/podcastdetail/9848
When your IoT Device Logs in as Admin, It’s too Late!
https://isc.sans.edu/diary/When+your+IoT+Device+Logs+in+as+Admin+Its+too+Late+Guest+Diary/32788
Apple Patches
https://support.apple.com/en-us/100100
Veeam Patches
SANS Internet Storm Center StormCast Thursday, March 12, 2026
Zombie Zip; How "Strengthening Crypto" Broke Authentication
https://isc.sans.edu/podcastdetail/9846
Analyzing "Zombie Zip" Files (CVE-2026-0866)
https://isc.sans.edu/diary/Analyzing+Zombie+Zip+Files+CVE20260866/32786
How "Strengthening Crypto" Broke Authentication: FreshRSS and bcrypt's 72-Byte Limit
https://pentesterlab.com/blog/freshrss-bcrypt-truncation-auth-bypass
SANS Internet Storm Center StormCast Wednesday, March 11, 2026
Windows, Fortinet, Adobe, and Zoom Patches
https://isc.sans.edu/podcastdetail/9844
Microsoft Patch Tuesday, March 2026
https://isc.sans.edu/diary/Microsoft+Patch+Tuesday+March+2026/32782
Fortinet Updates
https://fortiguard.fortinet.com/psirt
Adobe Updates
https://helpx.adobe.com/security.html
Zoom Update
https://support.zoom.com/hc/en/article?id=zm_kb&sysparm_article=KB0061222
Catch up on recent editions of NewsBites or browse our full archive of expert-curated cybersecurity news.
Free Virtual Summit | SANS Leadership Summit Solutions Track 2026 | Tuesday, March 17, 2026, at 10:00 AM ET.
Webinar | Don't Trust AI – Verify It: A Practical Discussion About AI in the SOC | Wednesday, March 18, 2026, at 1:00 PM ET.
Webinar | The Multicloud Blueprint: Architecting Security for the AI Era | Tuesday, March 31, 2026, at 1:00 PM ET.
Webinar | The Next Generation of Data Security | Tuesday, April 14, 2026, at 1:00 PM ET.