Oracle has released a patch for a severe vulnerability in its E-Business Suite (EBS) that could allow unauthenticated attackers to remotely access sensitive configuration data.
The flaw carries a high severity rating with a CVSS score of 7.5.
“If successfully exploited, this vulnerability may allow access to sensitive resources,” Oracle said in its recent advisory.
Recent attacks highlight ongoing risk for Oracle EBS users
CVE-2025-61884 poses a threat to enterprises running Oracle E-Business Suite, which supports essential functions like finance, manufacturing, and supply chain management.
If exploited, the flaw could let attackers skip authentication entirely and access sensitive business data.
The patch announcement follows a recent wave of extortion emails sent to executives at dozens of organizations, claiming that threat actors had stolen data from their EBS instances.
A different vulnerability, CVE-2025-61882, was likely exploited in that attack.
How the Oracle EBS vulnerability works
According to Oracle’s disclosure, the vulnerability resides in the Runtime UI of Oracle Configurator, a module used to manage product and service configurations within EBS.
It can be exploited remotely over HTTP — without authentication or user interaction — making it especially dangerous for internet-facing deployments.
The issue stems from an authentication bypass in how the Configurator Runtime UI validates user sessions.
Successful exploitation could allow attackers to retrieve configuration or system data without credentials. Because it primarily impacts confidentiality, Oracle has classified the vulnerability as a potential data exfiltration vector rather than a denial-of-service risk (DoS).
Oracle rates the flaw as network-accessible and of low complexity, meaning that attackers can exploit it without requiring privilege escalation or insider access.
Building a layered defense
Effective response requires not only immediate fixes but also strategic improvements to access control. Organizations should start with the following key steps:
- Apply patches: Install the latest patch to ensure all systems are up to date.
- Harden legacy systems: Migrate from unsupported or outdated versions and apply configuration hardening baselines to reduce exposure.
- Restrict and segment access: Limit HTTP and network access to administrative interfaces through segmentation, VPN restrictions, and firewall rules aligned with zero-trust principles.
- Monitor and log activity: Enable detailed logging and alerting for unusual authentication or HTTP activity.
- Review credentials and integrations: Enforce least privilege and multi-factor authentication (MFA) for admin accounts, and audit all connected APIs, middleware, and third-party integrations.
- Strengthen resilience and response: Conduct regular vulnerability scans, maintain secure offline backups, and update incident response plans to address enterprise resource planning (ERP) specific threats.
By combining immediate remediation with long-term access control improvements, organizations can better defend against evolving threats
While Oracle has not stated if it has observed active exploitation, organizations are encouraged to act preemptively.
Previous zero-day exploitation of CVE-2025-61882 led to data theft campaigns linked to financially motivated groups such as FIN11, which has previously leveraged the Cl0p ransomware in large-scale supply chain attacks.
ERP platforms remain prime targets for threat actors
The recent vulnerability of Oracle EBS highlights the persistent challenges in securing complex enterprise resource planning (ERP) systems.
These platforms — often containing vast stores of financial, operational, and customer data — remain high-value targets for attackers seeking maximum leverage.
Even with robust patch management programs in place, zero-day vulnerabilities and delayed updates can still leave organizations vulnerable.
As threat actors increasingly exploit ERP systems for data extortion rather than disruption, maintaining rigorous patch hygiene and continuous monitoring remains essential.
Editor’s note: This article first appeared on our sister publication, eSecurityPlanet.com.
Speaking of vulnerabilities, did you know that a recent investigation found that OpenAI models can be tricked into handing out instructions for creating explosives and chemical weapons? OpenAI said they addressed the issue, but anxiety lingers.