October is Cybersecurity Awareness Month, a time when organizations are encouraged to evaluate their cybersecurity programs and raise awareness about data security best practices.
This October, a wave of major cybersecurity incidents brought a stark reminder of why that work matters.
In what some pundits are calling “Black October,” organizations faced a series of significant security challenges, including critical vulnerabilities in PHP and a compromised release of a widely used open-source AI infrastructure tool.
While these incidents were not necessarily related, their convergence raises important questions about whether organizations across industries are equipped to identify, prioritize, and respond to modern threats.
Among the biggest headlines in “Black October” was the discovery of a pair of critical vulnerabilities in PHP, a widely used programming language that powers more than half of the world’s websites. Dubbed PHP2Shell, the flaws could allow unauthenticated remote code execution, potentially compromising vulnerable servers.
The vulnerabilities were particularly notable because artificial intelligence played a role in their discovery.
“AI has shrunk the gap between a flaw existing and a flaw being exploited,” said Kevin Lewis, senior consultant at BARR Advisory. “The PHP2Shell bugs sat unnoticed in PHP for over a decade before AI-assisted research found them, and organizations that couldn’t patch right away were told to pull their sites offline.”
The discovery underscored the growing role of AI in cybersecurity research—and the challenges it creates for practitioners. As AI tools make it easier to identify previously unknown vulnerabilities, organizations may have less time to assess their exposure and implement effective mitigation measures.
“Annual risk reviews and 30-day patch windows don’t hold up when discovery and exploitation are automated,” Lewis said. “AI tools are also a new attack surface. Every agent or integration employees connect is another set of credentials and data access to govern.”
The software supply chain was also targeted. A compromised release of LiteLLM, a popular tool for connecting applications to large language models, put organizations that installed the affected version at risk of credential theft, including the exposure of sensitive API keys and other secrets.
Other security incidents in October included major bugs in Node.js and GitLab, serious Linux kernel vulnerability disclosures, and an incident involving eMudhra that exposed weaknesses in the digital certificate infrastructure organizations rely on to establish trusted connections.
Taken together, these events demonstrate how vulnerabilities can extend well beyond one application. A weakness in one part of the technology supply chain can have implications for all the businesses, customers, and systems that depend on it.
While you can’t prevent every incident, you can build resilience in your organization by strengthening the practices that help you manage risk and respond to incidents. Here’s what to prioritize:
Effective vulnerability management starts with visibility. Organizations should maintain an up-to-date inventory of their software, applications, and integrations.
“Know what software and versions you run, and which vendors and AI tools hold your data or keys,” Lewis said. “You can’t patch what you don’t know you have.”
Long-lived API keys and broadly scoped credentials can give attackers access to sensitive systems and data long after an initial compromise. Where possible, replace them with short-lived, narrowly scoped credentials. Lewis also recommends establishing usage monitoring and spending alerts to help identify suspicious activity quickly.
“Set usage and spend alerts so abuse shows up in minutes, not on the invoice,” he said.
Traditional patching schedules may not be sufficient when vulnerabilities can be discovered and exploited at unprecedented speeds. Organizations should establish an emergency patching process with clear escalation criteria and pre-approved procedures to isolate systems that can’t be fixed quickly, Lewis advised.
“Some of the most damaging October incidents started at a vendor, not the victim,” Lewis said. “At least nine AI developer platforms were breached, and attackers went straight for the API keys customers had stored there, with stolen keys being abused within 30 minutes.”
A vendor’s SOC 2 report and other compliance credentials can provide valuable assurance, but they do not tell the entire story.
“Vendor reviews need to cover what access and secrets each vendor holds, and how fast you can revoke them,” Lewis said. “It’s not enough just to check whether they have a SOC 2 report.”
Organizations should establish clear procedures for responding when a vendor is compromised, including identifying affected integrations, determining which credentials need to be rotated, and assigning responsibility for taking action. Regular exercises can help ensure teams know what to do before an incident occurs.
“Rehearse your incident response plan, and know who rotates which keys and how fast,” Lewis said.
The events of Black October demonstrate that cybersecurity cannot be a static exercise. As AI accelerates vulnerability discovery and creates new opportunities for attackers, organizations must continually evaluate whether their security practices can keep pace.
Is your cybersecurity program ready to adapt to the changing threat landscape? Contact us today for a free consultation with our experts.