The Gap Between Disclosure and Attack Is Now Measured in Hours

Hourglass with sand running out, representing the shrinking time to patch vulnerabilities
Photo by Nathan Dumlao on Unsplash

For years, a common rule of thumb in IT was simple: install critical security patches within 30 days. That rule was written for a world where attackers needed weeks to figure out how to use a new vulnerability. That world is gone.

Today, the time between a vulnerability becoming public and attackers using it is often measured in days, frequently in hours, and sometimes it’s negative, meaning exploitation starts before the public even knows the flaw exists. For a small business still running on a monthly patch cycle, that gap is where the risk lives.

What the Numbers Show

Several major threat reports this year point in the same direction:

  • Google’s Mandiant M-Trends 2026 report estimated the average time to exploit a vulnerability at negative seven days, meaning that on average, exploitation begins a week before a patch exists (via the Cloud Security Alliance).
  • Rapid7’s Q3 2026 threat update found high-severity vulnerabilities were exploited an average of three days after public disclosure, with 80 percent of tracked exploitation happening within 14 days (Rapid7).
  • Industry research cited by DefenderMate found that about 28 percent of exploited vulnerabilities were weaponized within 24 hours of disclosure.

A “negative” average can sound strange. It doesn’t mean every flaw is attacked before it’s announced. It means enough of them are, often by well-resourced groups, that the average tips below zero. Either way, the practical message for business owners is the same: by the time you read about a critical vulnerability in the news, someone may already be using it.

Why Attacks Got So Fast

Three things changed at once.

AI made exploit development cheaper. Kaspersky’s Sergey Lozhkin told TechRadar that AI-assisted analysis makes it easier to turn a patch or advisory into a working exploit, so more attackers can act within hours or even minutes of technical details coming out (TechRadar). Researchers behind this month’s PHP2Shell discovery said AI tools compressed months of specialist work into days.

Patches themselves became roadmaps. When a vendor releases a fix, attackers can compare the old and new code to see exactly what changed, which points straight at the flaw.

Scanning became effortless. Attackers can sweep the entire internet for vulnerable systems in a matter of hours, so even small, obscure businesses get found quickly.

PHP2Shell Shows the New Timeline

This month’s PHP2Shell vulnerabilities are a textbook case. Details leaked before the official announcement. Amazon saw exploitation attempts four days before public disclosure. Fixes shipped on October 21, and CISA confirmed active exploitation five days later. A business that planned to “get to it in next month’s maintenance window” would have been exposed for weeks during the most dangerous period.

Meanwhile, Defenders Are Getting Slower

On the other side of the equation, organizations are struggling to keep pace. According to CISA, citing Verizon’s Data Breach Investigations Report, organizations fully remediated only 38 percent of known exploited vulnerabilities in 2024 (Paubox).

The math is unforgiving. If attackers typically start within three days and your patch cycle runs on a 30-day schedule, you’re potentially exposed for nearly four weeks on every critical, internet-facing flaw. Multiply that across a year’s worth of vulnerabilities, and it’s easy to see why “we patch monthly” is no longer a security strategy on its own.

What Changes for a Small Business

None of this means a small business needs a 24-hour security operations center. It means a few practical shifts:

  • Automate routine patching for laptops, browsers and common software, so the everyday updates happen without anyone remembering to do them
  • Create an emergency patch lane for critical, actively exploited flaws on internet-facing systems, separate from the monthly routine
  • Know what you have. You can’t patch quickly if you don’t know which systems run the affected software
  • Shrink your internet exposure. Firewall admin pages, remote desktop and old VPNs shouldn’t be reachable from the open internet if they don’t need to be
  • Add monitoring and endpoint detection, because some exploits will land before any patch exists, and someone needs to notice

A Simple Emergency Patch Playbook

  1. Someone watches. Assign responsibility for tracking vendor advisories and CISA’s exploited vulnerabilities list.
  2. Define “emergency.” For example: internet-facing, actively exploited, and capable of giving an attacker control or data access.
  3. Pre-approve the process. Decide in advance who can authorize an out-of-cycle change, so nobody waits for a meeting.
  4. Test quickly, not perfectly. A short sanity check on one system is better than a week of waiting.
  5. Deploy and verify. Confirm the fix actually applied.
  6. Check for compromise. If the flaw was exploited before you patched, look for signs that someone already got in.

Exploitation Speed Questions

How quickly are vulnerabilities exploited now?

Often within days of disclosure, and sometimes within hours. Rapid7 measured an average of three days for high-severity flaws in Q3 2026, and some are exploited before they’re even announced.

Is monthly patching still good enough?

For routine updates, a monthly cycle can work. For critical, actively exploited flaws on internet-facing systems, it leaves far too long a window. Those need an emergency process measured in days.

Why is AI making attacks faster?

AI tools help attackers analyze patches and advisories and build working exploits much faster than before, which shrinks the time defenders have to respond.

The Takeaway

The patch window hasn’t just shrunk. For the most dangerous vulnerabilities, it’s essentially closed. Businesses that hold up best keep routine updates automated, treat critical internet-facing flaws as emergencies, and have monitoring in place for the cases where attackers move first. If your plan for a critical vulnerability is “we’ll handle it in the next maintenance window,” that plan was built for a threat landscape that no longer exists.

Cloud9 Tech Solutions helps businesses close that gap with proactive network and server management, automated patching, and continuous monitoring as part of our security stack.

If you’re not sure how long a critical vulnerability would sit unpatched in your environment, that’s worth finding out before an attacker does. Schedule a free security assessment with Cloud9.