Cybersecurity professionals are enduring what some are calling "Black October," a month already marked by dozens of major cyberattacks and AI-identified vulnerabilities across the internet. The surge has prompted warnings from industry figures and governments, along with practical advice for ordinary users seeking to reduce their exposure.
Critical PHP flaws trigger global patching race
When critical vulnerabilities in PHP—the programming language used by around 70% of websites—were uncovered with the help of AI tools and disclosed by the PHP Foundation on Oct. 21, a race to patch servers worldwide began. Cloud services provider Cloudflare reported millions of attempts to exploit the vulnerability after disclosure, following thousands of attempts in the preceding hours. Some countries, including France and Finland, temporarily pulled access to certain websites, while the US National Security Agency warned organizations to patch the flaws immediately or take affected systems offline.
"We are at Security Armageddon now," said NEAR founder Illia Polosukhin. Anthropic's chief security adviser Jason Clinton warned that "as bad as things have gotten in the past 4 days, they are going to get a lot worse before they get better."
Citizens of the United Kingdom are being urged to "be ready" for an emergency, such as severe weather or cyberattacks, that could cut off access to critical infrastructure such as water, electricity or the internet. US military installations have been told to prepare for 90 days of independent operation under the FORTRESS initiative.
Recent incidents highlight shared infrastructure risks
In the past week or so, cloud platform Railway disclosed that an attacker accessed and decrypted customer credentials and API keys, apparently targeting AI providers. Medical testing giant Quest Diagnostics confirmed a breach of private patient health data, while Indian security firm eMudhra was found to have eight digital certificate authorities with weak keys that could potentially have let attackers impersonate trusted websites. Code hosting platform GitHub experienced a major service disruption that highlighted how heavily the software world relies on shared infrastructure.
GitHub's senior director of security, Austin Devers, said such incidents are a "clear reminder that reliability requires constant work," because modern development is "so connected."
Jonathan Riss, an OSINT and blockchain intelligence analyst at CertiK, told Magazine: "What concerns us most is how a single weakness in shared infrastructure can expose many organizations simultaneously." He added that AI's ability to find vulnerabilities is not matched by its ability to fix them: "Finding vulnerabilities faster does not automatically give us the capacity to resolve them faster."
AI lowers barrier for attackers
One reason the problem is worsening is that AI has changed the economics of cybersecurity. What once required nation-state resources and sophisticated hackers can now be achieved by a teenager in a basement somewhere, according to the source article. AI entrepreneur and investor Matt Shumer said AI enables non-technical users to find weak spots and exploit them, leaving defenders "racing to keep up."
"Every weak spot still has to be fixed, tested and installed on millions of systems run by people who are already stretched thin," Shumer said. Devers said AI-assisted security tools are "dramatically" lowering the barrier to "uncovering complex, deep-seated flaws in widely used codebases."
Practical steps for ordinary users
Despite the grim outlook, there are practical steps ordinary people can take to stay safer online during this volatile period.
Shumer recommends staying off sites you don't know, and definitely not entering passwords or card numbers into unfamiliar sites "until this blows over," since smaller sites are the "least likely to have patched." He also suggests deleting any apps you don't use and unplugging smart gadgets you don't need. Keep some cash at home and enough groceries for a few days "if you want to be extra careful."
Devers says the best defense is "locking down personal access points and keeping overall footprints small." Turn on multi-factor authentication (MFA) across all accounts, ideally using an authenticator app or hardware security key rather than text messages. "This acts as a circuit breaker and stops hackers from logging into your account even if the underlying software is compromised," he said. Never reuse passwords either: "Create strong, unique passwords for every site, and periodically disconnect old app permissions or third-party connections you no longer use." Keep your devices and web browsers updated automatically. "While platform teams work to fix cloud-level security issues on the backend, keeping your devices secure ensures hackers can't use your personal hardware as a back door into your accounts," Devers added.
Self-custody debate resurfaces
The alarming rise in security incidents has caused some crypto holders to consider alternatives, and some experts to advise against self-custody. Others, such as Bitcoin advocate and Casa chief security officer Jameson Lopp, remind users that giving up self-custody defeats the whole purpose. "You can only benefit from Bitcoin's unique properties when you hold your own keys. [...] Giving up your keys doesn't make the risk go away," he said. "It moves the risk somewhere you can't see or control."
Riss recommends separating small amounts you intend to use into a hot wallet from longer-term holdings in a hardware wallet, and to "actually read what you're signing rather than approving blind."
While these precautions can reduce vulnerability to attack, everyday users can only lock down local issues and have little ability to affect wider problems. You can "reduce blast radius," Riss said, "not prevent infrastructure compromise." So while users can improve their cyber hygiene, ultimately "the structural burden sits with platforms and providers."
