News

Majority of Data Breaches Go Unnoticed, Study Finds

Verizon Business reports that more than half of the data breaches on enterprise systems go undetected and are caused by general negligence and lax security.

More than half of the data breaches on hundreds of enterprise systems go undetected and are caused by general negligence and lax security, a report by Verizon Business revealed this week.

Verizon's 2008 Verizon Business Data Breach Investigations Report looked at some 500 cases between 2004 and 2007 where data were breached, resulting in more than 230 million compromised records.

The study revealed that 66 percent of the data breaches occurred due to incompetence and weak system fortitude. At least 75 percent of breaches evaded detection, with weeks, months and even years passing between incursion and discovery in 63 percent of the cases studied.

The reason for this, according to the study, is that "[f]irstly, and perhaps most obviously, criminals do not want to be discovered. They have great financial incentive to retain access to corporate systems for as long as possible and will go to great lengths to ensure their activities remain under the radar. Secondly, and perhaps most importantly, organizations simply are not watching."

Among some of the more startling findings is that 90 percent of all the hacks could have been avoided with security measures that are basic and "reasonable."

Verizon, as well as other security experts responding to the report, zeroed in on several characteristics of breach-prone environments. Among those is the fact that vulnerable systems often hold tens of thousands -- and sometimes millions -- of records. If a hacker gets in, they can do any number of things with individual records with little chance of detection.

Vulnerable systems also lack a viable breach notification process. It's one thing if a virus is uploaded; a user or administrator would probably notice. But data theft is quiet. Sometimes, files aren't even extracted or downloaded, but are merely copied to another location. It also gives the hacker the advantage of anonymity.

Recent high-profile cases like those at Hannaford Bros., TJX Cos. and, most recently, the Walter Reed Army Medical Center illustrate an endemic flaw in the policies, procedures and system integrity at some of the country's larger and well-known institutions.

"Hackers are really coming from all sides and from many vectors with, for the most part, clear objectives," said Michael Gavin, a security strategist at Boston, Mass.-based Security Innovation, a consulting firm that conducts annual and quarterly security audits for enterprise clients. "What's sensitive should be firewalled or put under the watchful eye of someone who knows what they're doing. Monitoring of how systems and processes work and are protected doesn't hurt either."

About the Author

Jabulani Leffall is an award-winning journalist whose work has appeared in the Financial Times of London, Investor's Business Daily, The Economist and CFO Magazine, among others.

comments powered by Disqus

Featured

  • Hands On with GitHub Copilot App Technical Preview: Turning a Blazor Issue into a PR

    GitHub's brand-new Copilot desktop app, in technical preview, handled a small Blazor issue from planning through pull request creation, but the hands-on test also showed why developers still need to verify agent work in the running app before merging.

  • At Build 2026, Microsoft Sets Up Windows as an OS for AI Agents

    Microsoft's Build 2026 Windows developer announcements point to a broader platform strategy for agentic AI, spanning terminal workflows, local models, app-building skills, Cloud PCs and operating system-level containment.

  • Slammed by Copilot Usage-Based Billing on Day 1, Facing $180 Bill for June

    A journalist using GitHub Copilot Pro details how a broken editorial workflow on day one of usage-based billing led to runaway token consumption, a projected $180 monthly bill, and practical tactics for cutting AI credit burn.

  • AdaBoost.R2 Regression Using C#

    AdaBoost.R2 regression works by building an ensemble of decision trees, training them on reweighted data, and combining their predictions with a weighted median, while also showing how parameter choices affect accuracy and overfitting.

Subscribe on YouTube