Artificial intelligence could trigger a new wave of cybersecurity emergencies as increasingly capable AI tools make it faster to discover software vulnerabilities and potentially exploit them. According to a Reuters analysis published on October 23, 2026, cybersecurity experts are warning that businesses may have less time to respond to newly discovered security flaws. The recent critical PHP vulnerabilities demonstrate why this matters for website owners, e-commerce businesses, and companies relying on third-party infrastructure.
On October 23, 2026, Reuters published an analysis based on interviews with seven cybersecurity experts examining how artificial intelligence is changing the discovery and exploitation of software vulnerabilities.
According to research from cybersecurity intelligence company VulnCheck, the number of reported software vulnerabilities has doubled year over year, partly driven by increasingly advanced AI tools.
AI can help researchers analyze large codebases, identify potential weaknesses, and evaluate whether discovered flaws could be exploited.
Tasks that previously required extensive manual investigation can now be accelerated with AI-assisted tools.
However, these capabilities are not limited to cybersecurity professionals working to protect software.
Attackers may also use AI to discover previously unknown vulnerabilities, develop exploitation techniques, and automate parts of their operations.
The concern is that businesses could face a shrinking window between the discovery of a critical vulnerability and attempts to exploit it.
VulnCheck reported approximately 100% year-over-year growth in published CVEs, which are standardized identifiers assigned to publicly disclosed cybersecurity vulnerabilities.
The company associates this increase with a combination of AI-assisted discovery and greater capacity among organizations authorized to assign CVE identifiers.
This distinction matters.
A growing number of CVEs does not necessarily mean that software is becoming twice as insecure. Some of the growth may reflect researchers finding previously undetected flaws or improvements in vulnerability reporting.
Likewise, a doubling of reported vulnerabilities does not mean that successful cyberattacks have doubled.
Not necessarily.
Although AI can accelerate the discovery of software vulnerabilities, researchers have not established that AI-discovered flaws are more likely to be exploited than vulnerabilities identified through traditional methods.
In its July 2026 State of Exploitation report, VulnCheck analyzed 1,061 vulnerabilities attributed to AI-assisted discovery.
Only 14 of those vulnerabilities, approximately 1.3%, had confirmed evidence of exploitation in real-world attacks at the time of the analysis.
This does not mean that the remaining vulnerabilities are harmless or will never be exploited.
It demonstrates the importance of distinguishing between identifying a vulnerability and successfully using it in an attack.
The same report also found that the median time between CVE publication and confirmation of known exploitation fell from 120 days in 2025 to 80 days in the first half of 2026.
This measurement applies to the vulnerabilities included in VulnCheck’s known-exploitation analysis, rather than all disclosed vulnerabilities.
VulnCheck also observed that the number of vulnerabilities exploited within 31 days of disclosure remained relatively stable despite the substantial increase in published CVEs.
For businesses, the critical issue is not simply how many vulnerabilities are discovered, but how quickly genuinely dangerous ones are identified, prioritized, and addressed.
The critical PHP vulnerabilities disclosed on October 21, 2026, offer a recent example of how serious software flaws can create widespread disruption.
The vulnerabilities, identified as CVE-2026-109618 and CVE-2026-109619, could allow remote code execution on affected servers under certain conditions.
PHP is a server-side programming language used by a substantial number of websites, including platforms built with WordPress, Magento, and other content management systems.
Because the vulnerabilities affect PHP itself, updating WordPress or its plugins does not resolve the underlying problem.
Website administrators and hosting providers were urged to apply emergency security updates.
The disclosure also prompted precautionary shutdowns of certain websites and online services while their operators assessed the risk.
According to Reuters, the vulnerabilities were discovered with the assistance of AI tools, highlighting the role that artificial intelligence is beginning to play in identifying long-standing security weaknesses.
The incident illustrates how quickly security teams may need to respond when vulnerabilities affect technologies used across a large portion of the internet.
Matthew Prince, CEO of Cloudflare, told Reuters that AI-powered vulnerability discovery could expose serious weaknesses in software that has supported critical internet infrastructure for years.
He warned that the coming years could involve frequent cybersecurity emergencies as researchers and attackers uncover more previously undetected vulnerabilities.
The concerns extend beyond PHP.
Reuters also discussed recent security incidents and vulnerabilities involving Node.js, Plex, LiteLLM, and Railway.
These incidents differ in nature and should not all be attributed directly to AI.
However, they illustrate the complexity of the modern software ecosystem, where organizations often depend on numerous applications, libraries, integrations, and external providers.
At the same time, there is a potential long-term benefit to faster vulnerability discovery.
Identifying and fixing previously unknown weaknesses could ultimately make software more secure.
AI may therefore increase the immediate operational pressure on cybersecurity teams while helping improve software security over time.
For website owners, the biggest challenge may be the need to respond to critical security issues outside their regular maintenance schedules.
Many businesses approach website updates as routine tasks performed weekly or monthly.
That approach remains useful for ordinary maintenance, but it may be insufficient when a critical vulnerability has been publicly disclosed and active exploitation is possible.
Businesses should ensure they have the processes and technical support required to assess urgent threats and respond appropriately.
Several areas deserve particular attention.
Organizations should keep track of security advisories affecting the technologies their websites rely on.
This includes content management systems, plugins, server software, libraries, and third-party integrations.
Not every vulnerability requires emergency action, but high-severity issues affecting publicly accessible systems should be evaluated promptly.
Businesses should know which technologies and software versions are running on their websites.
This information helps determine whether a newly disclosed vulnerability affects their infrastructure.
Without an accurate understanding of the technology stack, identifying exposure can take longer than necessary.
Website owners should understand who is responsible for maintaining each part of their infrastructure.
Depending on the hosting setup, responsibility may be divided between an internal development team, an external agency, a managed hosting provider, and a server administrator.
When a serious security issue emerges, uncertainty about ownership can delay necessary action.
Critical patches sometimes need to be applied outside normal maintenance windows.
Businesses should have access to reliable backups, a process for testing important website functionality, and a clear procedure for implementing urgent security changes.
For e-commerce websites, testing payment processing, shopping carts, and integrations is especially important.
Installing a security patch addresses the vulnerability, but it does not establish whether an attacker previously attempted to exploit it.
Server logs, application logs, WAF events, and security monitoring should be reviewed when appropriate.
If evidence of compromise is found, additional incident-response measures may be required.
Recent events involving PHP vulnerabilities demonstrate why ongoing technical maintenance and security monitoring are important.
At Delante, our development team has already taken action to assess potential risks affecting clients covered by our comprehensive development support.
As part of these efforts, our developers reviewed server configurations, checked relevant software versions, assessed potential security exposure, and increased monitoring.
For environments where Delante is responsible for server administration, our team directly implemented the necessary updates and security patches.
Where infrastructure is managed by external hosting providers or administrators, we contacted the responsible teams to coordinate updates and verify the status of security measures.
This distinction is important because responsibility for server-level maintenance varies between clients and hosting arrangements.
Our objective is to help clients understand whether a threat affects their website, identify the appropriate next steps, and coordinate technical actions within the scope of our services.
Businesses looking to assess their current website security can explore our WordPress Cybersecurity Audit.
For companies that need ongoing technical support, our WordPress Administration service includes website maintenance, security monitoring, backups, and scheduled updates.
A successful cyberattack can affect much more than a website’s server environment.
Depending on the nature of the compromise, attackers may inject malicious code, modify website content, create spam pages, or redirect visitors to unauthorized destinations.
These changes can negatively affect the user experience, website availability, and brand reputation.
In more serious cases, search engines may identify compromised websites as unsafe, discouraging users from visiting them.
E-commerce businesses may also experience interrupted transactions, lost sales, and damage to customer trust.
However, the existence of a security vulnerability does not automatically cause a website to lose Google rankings.
SEO consequences depend on whether a compromise actually occurs and how it affects the website.
For businesses, maintaining technical security should therefore be part of a broader strategy for protecting website performance and visibility.
You can learn more in our guide to SEO Cybersecurity and Protecting Your Website From Attacks.
If your website has already been compromised, our article My Website Has Been Hacked. What Should I Do? explains the steps worth considering.
As AI-assisted vulnerability discovery becomes more widespread, businesses may need to review how they manage software updates and cybersecurity responsibilities.
The challenge is not necessarily to patch every newly disclosed vulnerability immediately.
Instead, organizations need to distinguish between issues that present a significant risk and those that can be addressed through normal maintenance processes.
This requires reliable information about software vulnerabilities, knowledge of the organization’s technical infrastructure, and clear responsibility for implementing fixes.
Companies working with external hosting providers and development teams should also establish effective communication procedures.
When a critical vulnerability is disclosed, everyone involved should understand who will investigate the issue, decide on necessary action, deploy patches, and confirm that the affected systems are secured.
The growing capabilities of AI could make this operational readiness even more important.
AI is increasingly influencing how software vulnerabilities are discovered, evaluated, and potentially exploited.
Experts interviewed by Reuters warn that this development could result in more frequent cybersecurity emergencies and shorter response windows for affected organizations.
However, the available research does not demonstrate that the number of successful cyberattacks is growing at the same rate as vulnerability disclosures.
VulnCheck’s findings suggest that most vulnerabilities discovered with AI assistance had no confirmed evidence of real-world exploitation at the time of its analysis.
At the same time, the potential for attackers to use increasingly capable AI tools remains an important concern.
For website owners, the main takeaway is clear: cybersecurity is no longer just about regular software updates. It also requires risk-based prioritization, reliable monitoring, and the ability to respond quickly when serious vulnerabilities are discovered.
As AI continues to evolve, businesses that understand their infrastructure and maintain effective security procedures will be better positioned to manage emerging risks.
