Howdy!
Welcome back to the InfoSec GRC Brief from SaltyCloud, your go-to source for all the latest insights on governance, risk, and compliance (GRC) in information security.
Let’s get into it.
Cyber Incidents & Risk Implications
FAA Halts Departures Nationwide After Communication Problems
On November 13, the Federal Aviation Administration (FAA) halted departures at airports nationwide after communication problems affected multiple Air Route Traffic Control Centers.
The ground stop held aircraft at their departure airports. Airlines said flights already in the air would continue, while the FAA said inbound international flights would be handled individually.
The FAA also said departures were halted until further notice while its technicians investigated. In a separate statement, the agency described a temporary telecommunications outage, said the ground stop was precautionary, and reported that every flight had been accounted for. An internal Alaska Airlines message similarly described communication and radar problems.
The disruption affected hundreds of flights, with 11,000 flights scheduled to depart U.S. airports after 4 p.m. Eastern that day. Meanwhile, as crews reach their daily duty-hour limits, prolonged stops can also compound delays.
The last nationwide ground stop occurred in January 2023, when a failure in the system used to distribute notices to pilots interrupted departures.
Microsoft 365 Outage Disrupts Security Tools
Microsoft’s November 12 outage continued into November 13, disrupting email and the security and compliance services connected to it.
In an earlier update, Microsoft attributed the disruption to an issue introduced during a network capacity upgrade that surfaced after a subsequent network change. The company described the incident as technical and unrelated to cyberattacks or AI agents.
But the affected services extended beyond Outlook and Exchange Online:
- Defender XDR portal pages could fail to load, while some Defender for Office 365 policy evaluation and enforcement experienced delays or failures.
- Intune device enrollment, check-ins, and compliance policies were intermittently affected.
- Purview disruptions included data loss prevention, auto-labeling, audit searches, eDiscovery searches and exports, and access to protected content.
Most core Teams calls, meetings, and chats stayed functional, while Exchange-dependent calendar features were affected. Direct access to files stored in SharePoint also remained available.
In its November 13 update at 2 p.m. Pacific, Microsoft reported sustained recovery and said most affected scenarios had been addressed. However, the company is still monitoring the backlog of email messages and dependent services before declaring the incident resolved.
Fireworks AI Breach Exposes Uber and Notion Data
Uber confirmed that a breach at AI provider Fireworks AI exposed driver information, including driver’s license details.
In its November 10 notification letter, Uber said a service provider reported compromised internal credentials on October 9. Uber then disconnected the provider and confirmed unauthorized access to Uber data on October 13.
Notion also disclosed that the incident exposed AI chat transcripts used by its Insights feature, which analyzes customers’ AI usage. Its November 12 advisory says fewer than 1% of workspaces were affected. But the assessed scope changed over several days:
- November 2: Fireworks alerted Notion to a dataset initially identified as synthetic testing data.
- November 3–4: Notion disconnected from Fireworks and deleted its data from the provider.
- November 6–9: Fireworks reported potential customer-data exposure, confirmed the affected dataset, and Notion notified customers.
Most affected transcripts lacked user or workspace labels and contained no tool-call results, but the conversations themselves could include identifying information.
On November 11, Notion subsequently found exposed credentials in user messages. The company revoked Notion API keys and notified other key issuers on November 12.
Notion has disabled Insights and stopped serving open-weight models through Fireworks, but its investigation remains open. Affected customers can review their exposed data inside their workspaces.
SignDrop Exploit Targets Windows Privilege Escalation
On November 10, a researcher using the name NightmareEclipse published SignDrop, an exploit that can turn ordinary local-user access into SYSTEM privileges on qualifying Windows installations.
The disclosure targets the Common Log File System (CLFS), a Windows logging component. When a privileged process such as Microsoft Defender opens a crafted log, CLFS can create an authentication file under SYSTEM privileges while retaining permissions associated with the original file. The exploit abuses that interaction to gain elevated access.
The researcher’s listed test environments were:
- Windows 11 26H2
- Windows 11 26H2 (Experimental)
- Windows Server 2025 (Experimental)
The exploit also requires CLFS authentication to be in learning or enforcement mode—and those conditions matter. For example, researcher Will Dormann independently reproduced the behavior, but only one of his tested machines qualified.
Microsoft said it was investigating but had no evidence of exploitation in the wild, and that Defender blocks potential exploitation attempts.
Researchers Report N-central Exploitation After Hotfix
Researchers reported exploitation of N-able N-central on November 12, just two days after the company released Hotfix 2.
The newly disclosed 4EverAble2Shell exploit can give an unauthenticated attacker root access when the remote monitoring and management platform’s Legacy API is enabled.
Huntback’s investigation traced the activity from public disclosure to attempted exploitation within hours. Researchers first saw probes against their decoys about four and a half hours after the proof of concept appeared, followed by requests that chained code execution and privilege escalation to install a cryptocurrency miner.
A public mining pool then showed 21 active machines whose addresses hosted N-central installations. Nineteen reported build 2026.4.1.26—the November 10 Hotfix 2 release.
At the time, N-able’s advisory said it had not seen evidence of compromise using the newly disclosed flaws. However, the company also said it was restricting functionality while developing further fixes and urged on-premises customers to disable Legacy API access.
The Hotfix 2 release notes also require customers to update both Agents and Probes, including when N-able manages the hosted server. N-able recommends updating Agents first to preserve uninterrupted communication between the components. So a hosted-server upgrade does not complete the customer-side component updates. However, Huntback’s findings show that Hotfix 2 alone did not prevent the reported attacks.
Regulatory & Compliance Updates
FedRAMP Details Vulnerability-Management Migration Evidence
On November 10, FedRAMP clarified the evidence cloud providers must submit as they migrate to its Vulnerability Detection and Response (VDR) and Vulnerability Evaluation and Reporting (VER) rules.
The December 7 migration deadline was established in June. The rules move vulnerability management beyond CVE/CVSS scores and fixed response windows toward context, reachability, and risk.
VDR also covers weaknesses beyond software vulnerabilities, including failures in security processes and outdated control statements. Providers must detect and address those weaknesses as part of operating the cloud service.
More specifically, Notice 0017 requires a progress report containing:
- A Mini-Security Decision Record with implementation status and explanations for each applicable rule
- Rule-specific corrective actions and timelines for required work that remains incomplete
- Sample machine-readable and human-readable vulnerability reports, each containing at least five examples of each vulnerability type
FedRAMP will collect the reports through an early-December Security Inbox Emergency Test. The anticipated response deadline is December 17 at 2 p.m. Eastern, subject to test notification. Providers planning completion after February 3, 2027, will be listed as Certified (In Remediation).
The general grace period ends March 7, but revocation and a one-year ban apply only when a provider has neither completed migration nor obtained a FedRAMP-approved corrective action plan. Still, extended grace requires an approved plan and agreement to a FedRAMP-coordinated red-team assessment.
Five States Join FTC Lawsuit Against Hims & Hers
On November 10, Arizona, Florida, Nebraska, Ohio, and Wisconsin joined the FTC’s lawsuit against Hims & Hers.
The complaint alleges violations of the FTC Act, the Restore Online Shoppers’ Confidence Act (ROSCA), and state consumer-protection laws.
The case was originally brought in July with Utah and California. The amended complaint suggests that the telehealth company shared sensitive health information with Meta, Snap, and other third parties despite its privacy promises.
The FTC’s original announcement describes two disclosure mechanisms:
- Customer lists supplied to advertising platforms
- Tracking technologies that automatically transmitted visitors’ website actions
Affected customers provided health information through online intake forms intended for medical-provider review before treatment.
The case also challenges billing and cancellation practices. The FTC alleges that customers could be charged or enrolled in subscriptions after completing intake forms but before a consultation to review and approve treatment. It also says that the cancellation option was hidden behind an “add/remove items” button.
The allegations remain pending in federal court.
CNIL Sanctions Public Bodies Over Security Failures
On November 9, France’s data protection authority CNIL announced ten sanctions against local authorities and a public establishment.
Following inspections of public bodies and their service providers, the CNIL issued fines ranged from €3,000 to €20,000. Under GDPR Article 32, the findings included:
- Remote administrator access without multifactor authentication
- Vendors with permanent network access or accounts shared by multiple employees
- Inadequate logging and network segmentation
- Unsupported server operating systems
- Weak passwords, passwords sent in plaintext, or storage that did not protect confidentiality
Some inspected authorities had suffered data breaches that appropriate safeguards could have prevented. However, several decisions also identified systems put into service without the security approval required by France’s General Security Framework.
Separate findings concerned missing data protection impact assessments. Examples included portals for managing children’s school activities that also processed banking details, and public-space video surveillance involving systematic monitoring at scale.
CNIL said the fines reflected the severity and number of failures, the data and people affected, each authority’s budget, and corrective measures already taken. Further enforcement proceedings remain underway.
GRC Strategies
United Restricts Unauthorized AI Agents
On November 11, United Airlines announced that it is restricting unauthorized third-party personal AI agents from accessing its digital platforms.
The restriction applies both directly and through travel agencies and aggregators. More specifically, the airline’s announcement identifies two related concerns:
- Customers sharing account credentials with AI assistants
- Assistants taking actions that result in unintended purchases, account changes, or use of MileagePlus miles
United is updating its booking, ticketing, and distribution policies to enforce the restrictions.
The airline also cited the need for human acknowledgment during flight check-in, describing how automated actions could conflict with FAA and Department of Transportation requirements.
For customers affected by unintended agent actions, United plans to update its Digital Trust Center with explanations of the restrictions.
Nvidia Changes Leadership Amid Board Investigation
On November 12, Nvidia announced that its president, CEO, and a director, Jensen Huang had resigned, effective immediately.
CFO Colette Kress is now the interim CEO, and Harvey C. Jones is the company’s first board chair. The leadership change followed the board’s November 8 announcement of an independent investigation into reports that Huang had been targeted by a foreign intelligence effort. At that point, the board said it had no reason to believe national security or Nvidia secrets had been compromised.
On November 10, independent directors named Joon H. Kim and Wachtell, Lipton, Rosen & Katz as outside counsel for the inquiry.
Meanwhile, Nvidia is considering internal and external candidates for a permanent CEO. In its transition announcement, the board said it expected operations, customer commitments, and the product roadmap to continue without interruption.
FAA Expands AI-Assisted Air-Traffic Planning
On November 10, the FAA announced that its Strategic Management of Airspace, Routes and Trajectories (SMART) system would expand to the airspace served by all 22 centers nationwide.
During the September rollout, the National Air Traffic Controllers Association reported that it had not participated in SMART’s design, testing, or implementation. According to the FAA:
- The cloud-based system combines 200 data streams, including weather, flight paths, traffic flow, and controller staffing.
- A Washington-based team analyzes and validates the data before sending scheduling recommendations to the Air Traffic Control System Command Center in Warrenton, Virginia.
- The Command Center then forwards those recommendations to local leadership for consideration.
After the initial launch, the FAA says SMART helped reroute 180 flights, saving an average of 10 minutes and nearly 80 gallons of fuel per flight.
NCSC Introduces Adversary-Simulation Provider Scheme
On November 11, the UK National Cyber Security Centre launched its Cyber Adversary Simulation (CyAS) scheme to help critical organizations obtain realistic tests of their cyber defenses.
The scheme assesses commercial providers against an NCSC-defined standard. More specifically:
- Providers first investigate a customer’s technology, exposure, and threat environment, then develop attack plans tailored to that organization.
- Testers use an adversarial approach to pursue agreed objectives and examine prevention, detection, and response capabilities.
NCSC says the methodology is not limited to a fixed set of documented tactics or commercially purchased threat intelligence. Reconnaissance and the customer’s actual environment shape the exercise instead.
Developed with regulators and lead government departments, CyAS complements the Cyber Assessment Framework and Cyber Resilience Audit scheme, giving oversight bodies another source of evidence about resilience across the sectors they supervise.
The launch is an initial minimum viable product intended to test demand and refine the methodology before a fuller operating model. In the meantime, NCSC has published a list of assured providers.
Regulatory Experts Recommend Planning Before Deadlines
Regulatory experts at the RAPS Regulatory Intelligence Conference in Barcelona described how companies can prepare for major legal changes before final requirements are available.
Bertrand Fournier, senior director for regulatory policy at Alnylam Pharmaceuticals, used the EU’s developing pharmaceutical legislation as an example. His team was already conducting gap analyses and impact assessments using pre-adoption documents, with plans to repeat that work against the final text.
Regulatory intelligence must account for the legislative, scientific, technological, and political environment alongside the company’s strategy, Fournier said—including plans to enter new regions.
Katy Thompson, director for clinical trials regulatory at IQVIA, described the cross-functional work needed to turn that intelligence into implementation. Quality assurance, legal, pharmacovigilance, and other teams should assess effects on their own processes, while IT helps build a system that tracks requirements, required documents, and timelines.
Ultimately, the findings should feed a roadmap that translates early intelligence into a working regulatory plan.
SaltyCloud Research
TPRM Maturity Checklist
Assess any third-party risk program with a checklist grounded in NIST CSF 2.0 and NIST SP 800-53, plus HECVAT for higher education.
Vendor Risk Assessment: Complete Guide [2026]
Learn about the vendor risk assessment process, security questionnaires, and assessment reports used to evaluate third-party risk.
The 18 CIS Controls: Complete Guide [2026]
See all eighteen CIS Controls and their safeguards—access management, vulnerability management, service-provider oversight, and penetration testing—all in one place.
Wrapping Up
The FAA and Microsoft outages showed how quickly a technical problem can disrupt essential operations. Microsoft’s incident also interrupted security and compliance tools, a reminder that the services an organization relies on to manage risk can be affected by the same disruption it is trying to recover from.
If you’re on LinkedIn, follow SaltyCloud for more insights and updates between newsletters. And, as always, feel free to reply to this email with comments, questions, concerns, ideas, or memes. We’re always listening.
Thanks for being part of our community and for letting us be part of your day. We’ll keep doing our part to deliver content that’s as essential as salt itself.
Wishing you resilience and security,
The SaltyCloud Research Team
This content is for informational purposes only and does not constitute legal or compliance advice. See our full disclaimer.
Dive into our research-backed resources–from product one pagers and whitepapers, to webinars and more–and unlock the transformative potential of powerfully simple GRC.
Learn More