Americas

  • United States

Asia

Oceania

by Dave Gradijan

Data-Leakage Prevention Tools Catch Errors, Not Theft

News
Sep 12, 20073 mins
Build AutomationCSO and CISO

Platforms that detect when sensitive corporate data is leaked are more effective against people making honest errors than they are against criminals trying to steal the data, says one analyst.

A small percentage of data that leaks from corporate networks (0.5 percent) is stolen by professionals whose efforts will evade detection by security products touted as data-leakage prevention tools, says Nick Selby, an analyst with 451 Group who spoke at the Security Standard conference Monday.

The products do catch data leaks, 98 percent of which are linked to an accident or stupidity and 1.5 percent that are caused by vengeful employees clumsily attempting to steal data, he says. “Data leakage is an antistupidity issue as much as it is a technology issue,” Selby says. “Most data-leakage products can’t discover activity by skilled insiders looking to steal.”

He says the products should be recast as tools that can help eliminate data breaches made in error, rather than those that are done intentionally.

Selby named Reonnex, Vontu, Onigma (bought by McAfee), Tablus (bought by EMC/RSA) and Port Authoritiy (bought by WebSense), among many others. He singled out these vendors because they have been the recipients of venture-capital dollars. He says the total investment in this technology amounts to $250 million. “That’s an awful lot of money,” Selby says, “and vendors, you’re not going to fix [the problem].”

The problem of keeping corporate data safe is complicated, he says, because most businesses don’t have any idea where all their data is, much less have it classified as sensitive or not. Leakage products can be useful because they crawl network databases, locating and categorizing data.

To accurately detect abuse, these products need to track behaviors of users. “It needs context. What did they do, why and have they done it before?” Selby says. If a device cannot determine whether access to certain data represents a threat, it should note that.

“It says, ’I don’t understand. Type in why you’re doing that,’” he says. At least that can alert an inadvertent data leaker to the error and leave a trail of explanations for auditors to check later should a leak prove malicious.

Selby projects that within 19 months, data-leakage technology will be blended in with other products that scan network traffic. He advises customers of such products to buy from vendors that offer more than just data-leakage products so they are more likely to be in business in a year or two when they need support.

By Tim Greene, Network World (US)