eSentire's Threat Response Unit (TRU) compiles the following weekly intelligence overview, providing cybersecurity leaders with expert analysis and insights on the most important events of the past week along with important security tips.
Recent Threat Intelligence Advisories
Noteworthy News
Critical Progress Kemp LoadMaster Vulnerability Under Active Exploitation (CVE-2026-8037)
Bottom Line: eSentire’s Threat Response Unit has identified attempted exploitation of the critical Progress Kemp LoadMaster vulnerability CVE-2026-8037. Organizations with vulnerable devices are recommended to apply the relevant security patches immediately.
On June 4th, 2026, Progress disclosed CVE-2026-8037 (CVSS: 9.8), a critical OS Command Injection Remote Code Execution (RCE) vulnerability in the Kemp LoadMaster application delivery controller. Kemp LoadMaster is a load balancer and Application Delivery Controller (ADC) deployed at the network edge in enterprise environments. The vulnerability allows an unauthenticated attacker to execute arbitrary commands on the LoadMaster appliance by sending a specially crafted request to the /accessv2 API endpoint when the API feature is enabled. Progress also disclosed a second vulnerability in the same security bulletin, CVE-2026-33691, which is addressed by the same patches.
On June 29th, 2026, WatchTowr Labs published a detailed technical analysis and functional Proof-of-Concept (PoC) exploit demonstrating this technique. On the same day that the PoC was released, eSentire's Threat Response Unit (TRU) identified exploitation attempts targeting CVE-2026-8037. In the cases observed by eSentire, the exploitation attempts matched WatchTowr’s PoC, but were not successful, as such, no post-compromise activity was identified.
eSentire Threat Intelligence Analysis:
LoadMaster's network edge position makes it a high-value initial access target. ADC appliances handle authentication flows, terminate TLS, and often have broad visibility into internal service infrastructure. Compromised appliances provide an attacker with both a network foothold and information about the internal environment.
CVE-2026-8037 was exploited in under one month from its initial disclosure and within 24 hours of the release of technical details. This timeline illustrates the limited patch window that is common with new vulnerabilities. Organizations must prioritize the rapid patching of newly released vulnerabilities, especially when technical details or PoC exploit code is released. Five Eyes intelligence agencies have recently warned that the patch window will continue to decrease, due to the widespread adoption of AI by threat actors.
eSentire's TRU has released an advisory on this topic and developed detections within MDR for Network. Known attacker infrastructure identified in observed exploitation attempts has been added to eSentire's Global Block List, providing automated blocking for MDR for Network customers. eSentire's Managed Vulnerability Service (MVS) will add relevant detection plugins as they become available. TRU is actively tracking this threat for additional indicators and detection opportunities. Organizations running affected LoadMaster versions must apply patches to GA v7.2.63.2 or LTSF v7.2.54.18 immediately.
Recently Disclosed NetScaler Vulnerability Exploited (CVE-2026-8451)
Bottom Line: The NetScaler ADC and Gateway vulnerability, CVE-2026-8451, is under active exploitation only days after it was publicly disclosed.
On June 30th, 2026, Citrix released security updates addressing multiple vulnerabilities in NetScaler ADC (formerly Citrix ADC) and NetScaler Gateway (formerly Citrix Gateway). Technical details for one of the vulnerabilities have been publicized and attacks targeting the vulnerability were identified the same day.
The confirmed exploited vulnerability is CVE-2026-8451 (CVSS: 7.5). It is a memory overread vulnerability in NetScaler’s SAML XML parser that allows an unauthenticated attacker to disclose arbitrary process memory on appliances configured as a SAML Identity Provider (IdP). In an attack scenario, the threat actor would send specially crafted login requests to the device without needing a username or password and trick it into leaking small fragments of its internal memory back in the response. That leaked memory can contain session data or other sensitive information.
WatchTowr researchers released a technical analysis for the vulnerability on June 30th, along with a Detection Artefact Generator that can be used for identification and remediation. Within hours of technical details becoming available, the security firm Lupovis reported observing exploitation attempts. Researchers noted that the payload structure closely matched the detection artifact generator released by WatchTowr, indicating that attackers were likely leveraging publicly available research to identify vulnerable targets. eSentire’s Threat Response Unit (TRU) has observed scanning and exploitation attempts for CVE-2026-8451, although they have not resulted in successful compromises.
Other vulnerabilities from this release that have not been identified as exploited include CVE-2026-8452, CVE-2026-8655, CVE-2026-10816, CVE-2026-10817, and CVE-2026-13474
eSentire Threat Intelligence Analysis:
The disclosure and rapid exploitation of CVE-2026-8451 highlights the persistent risks associated with Internet-facing Citrix NetScaler infrastructure and reinforces the trend of threat actors weaponizing newly disclosed edge-device vulnerabilities within hours of public release. From a threat landscape perspective, the most significant development is not the vulnerability itself but the speed at which threat actors operationalized public research. Within less than 24 hours of disclosure, security researchers observed attackers scanning the Internet for exposed NetScaler appliances and immediately delivering exploit payloads upon identifying potentially vulnerable systems. The observed activity demonstrates how quickly adversaries can leverage publicly available Proof-of-Concept (PoC) material to identify exposed targets, reducing the window available for defenders to patch and remediate vulnerable assets.
The vulnerability is particularly concerning because Citrix appliances continue to be attractive high-value targets. As network edge devices responsible for authentication, remote access, and traffic management, successful exploitation can provide attackers with access to sensitive session data, authentication material, or information that can facilitate subsequent intrusion activity. Previous CitrixBleed vulnerabilities were heavily exploited by threat actors to gain initial access to enterprise environments.
As exploitation of CVE-2026-8451 has been confirmed, it is critical that organizations with vulnerable devices apply the relevant security patches immediately. If patching cannot be performed immediately, administrators can temporarily disable SAML Identity Provider functionality where operationally feasible.
TRU assesses it is highly likely that both targeted and opportunistic exploitation attempts against Internet-facing NetScaler appliances configured as SAML IdPs will continue to increase over the coming weeks. Organizations that have not yet patched or disabled SAML IdP configuration should treat this as an urgent action item.
eSentire MDR for Network has detections in place to identify exploitation attempts targeting CVE-2026-8451. In addition, threat hunts have been performed and an advisory on the topic was released on July 3rd.
Fraudulent OpenAI Tenant Impersonates Security Company
Bottom Line: Threat actors have been identified abusing OpenAI’s organization invitation feature to trick users into submitting sensitive data that could be used for malicious purposes. The attack exploits the trust that users place in AI platforms.
Push Security has disclosed an attack targeting the company, in which threat actors attempted to trick employees into submitting data to an attacker controlled OpenAI organization. Multiple employees had received targeted email invitations to join an OpenAI organization named “Push Security Inc”. The invitations originated from OpenAI’s legitimate notification address and as such, passed all standard email authentication checks. The only technical indicator of malicious intent was a single warning line in the email body, noting that the inviting account’s email domain did not match the recipient’s corporate domain.
To investigate the attack, Push Security staff accepted one invitation. Accepting the invitation required a single click with no credentials or additional authentication. Once accessed, the fraudulent OpenAI organization showed that the attacker’s account was configured under the name of Push Security’s CEO. The account relied on a stolen credit card to pay for usage. With access to the organization, Push Security confirmed that no other employee had accepted the invitation. This attack has been dubbed the “poisoned tenant” technique.
eSentire Threat Intelligence Analysis:
Push Security assessed the attacker’s primary objective as passive data collection. Had employees assumed the organization was a legitimately provisioned company account and began using it, the attacker would have had persistent, passive access to all prompt content, usage logs, and API interactions. AI platforms are routinely used to process sensitive material including source code, internal documents, customer data, and strategic plans meaning the threat actor could gather significant amounts of sensitive data. As AI platforms expand to serve as orchestration hubs for business workflows, connecting to email, calendars, file storage, and cloud services, the downstream impact of an employee operating unknowingly within an attacker-controlled environment grows significantly.
Organizations should treat an employee joining any unverified AI platform as a security-relevant action carrying a risk profile comparable to installing an unverified application or authorizing an untrusted OAuth integration. User training should be updated to specifically address the poisoned tenant technique, as the sender email address is a legitimate AI platform email. An easily accessible and clear AI usage policy will help employees to understand what tools they can access and identify potentially malicious AI lures.
About the threat briefings:
The threat briefing is a weekly intelligence overview for non-technical users. The Threat Intelligence team at eSentire investigates, analyzes, and organizes the most important events of the past week along with important security tips and redistributes the findings for quick reading. The main goal of this briefing is to improve the overall awareness of readers regarding cyber security.
