Although many security experts predict that the most likely data breach threats organisations will face in 2013 include cloud exploits, mobile device attacks and all-out cyber war, “Verizon Data Breach Investigations Report” (DBIR) researchers have reached a far different conclusion: The most likely threats involve authentication attacks and failures, continued espionage and “hacktivism” attacks, Web application exploits and social engineering.
The findings of the researchers — members of the company’s RISK (Research Intelligence Solutions Knowledge) Team – are based on data that spans eight years and thousands of cases and is contained in the 2012 data breach report, released earlier this year.
”Many security experts are using anecdote and opinion for their predictions, whereas Verizon’s researchers are applying empirical evidence to help enterprises focus on what will be truly important in the coming year — and also what isn’t,” said Wade Baker, principal author of the DBIR.
“First and foremost, we don’t believe there will be an all-out cyber war, although it’s possible,” he said. “Rather, an enterprise’s 2013 data breach is much more likely to result from low-and-slow attacks.”
Verizon’s RISK team has identified the following most likely data threats:
- Topping the list – with a 90 percent change of probability — are attacks and failures related to authentication, including vulnerable or stolen usernames and passwords, which often represent the initial events in a breach scenario. “Nine out of 10 intrusions involved compromised identifies or authentication systems, so enterprises need to make sure they have a sound process for creating, managing and monitoring user accounts and credentials for all of their systems, devices and networks,” Baker said.
- Web application exploits which are most likely to affect larger organisations and especially governments, rather than small to medium-sized businesses. The chances of such attacks occurring are three in four, according to the data compiled by the RISK Team. “Given these odds, organisations that choose to take their chances and ignore secure application development and assessment practices in 2013 are asking for trouble,” said Baker.
- Social engineering, which targets people rather than machines and relies on clever — and sometimes clumsy — deceptions to be successful. “The use of social tactics like phishing increases by a factor of three for larger enterprises and governments,” said Baker. “It’s impossible to eliminate all human error or weaknesses from an organisation, but vigilance and education across the employee population help to control and contain such schemes.”
Baker also said that targeted attacks from adversaries motivated by espionage and hacktivism — breaking into a computer system, for a politically or socially motivated purpose — will continue to occur, so “it’s critical to be watchful on this front.”
In addition, the RISK team does not foresee the failure of an organisation’s cloud technology or configuration as being the root cause of a breach. However, an organisation’s service provider could inadvertently increase the likelihood of a breach by failing to take appropriate actions or taking inappropriate ones.
As for mobile devices, the Verizon researchers believe that lost and stolen – and unencrypted — mobile devices will continue to far exceed hacks and malware.
The RISK Team also projects that attacks on mobile devices by the criminal world will follow closely the push to mobile payments in the business and consumer world. “There’s a good chance we’ll see this shift in 2013, but our researchers think mobile devices as a breach vector in larger enterprises will lag beyond 2013,” Baker said.
Large organisations tend to pride themselves on their security strategy and accompanying plans, but the reality is that a large business is less likely to discover a breach itself than being notified by law enforcement. “And, if you do discover it yourself,” Baker said, “chances are it will be by accident.” He concluded: “Keep in mind that all of these breaches can still be an issue for enterprises. However, what we’re saying is that they’re over-hyped according to our historical data and are far less likely to factor into an organisation’s next breach than is commonly thought.”
