IBM Security announced its 2018 X-Force Threat Intelligence Index on April 4, revealing that the number of breached records in 2017 was actually less than what was reported for 2016.
According to IBM's analysis, 2.9 billion records were breached in 2017, down 25 percent from the 4 billion stolen customer data records in 2016. The decline in the number of reported data records that were stolen in breaches, however, is not an indication that cyber-crime has declined, but rather that cyber-crime has shifted.
"For this report, the number of compromised records was calculated based on our analysis of publicly reported data breaches," Michelle Alvarez, threat researcher at IBM X-Force and one of the lead authors of the report, told eWEEK. "Organizations are not required to report ransomware and because of this, there isn't a running list of every ransomware attack and there isn't a per record tally like there are for breaches.
"Basically, we're saying that you can't judge the state of cyber-crime by how many records are reported as breached; there are many other factors that aren't tracked," she said.
If the amount of records or devices compromised by ransomware were to be counted together with breached records, the total figure could be much higher than past years' reports, Alvarez said. For example, she said the WannaCry ransomware outbreak alone reportedly affected hundreds of thousands of devices, and some organizations were unable to recover the files due to the nature of the attack.
A key finding in the 2018 X-Force Threat Intelligence Index is that human errors in misconfigured cloud infrastructure accounted for approximately 70 percent of the total number of compromised records tracked by IBM X-Force in 2017.
"The majority of the reported misconfigured cloud incidents were from Amazon S3 buckets, which exposed large amounts of sensitive data without requiring authentication," Alvarez said.
There were multiple publicly reported incidents involving misconfigured S3 storage buckets in 2017, including ones from Alteryx that exposed information on 123 million Americans and one at Verizon impacting 6 million individuals.
IBM also reported a spike in the volume unexpected code injection attacks in 2017. Unexpected code injection was the mechanism for 79 percent of attacks against security clients monitored by IBM, up from 42 percent in 2016.
"Coin mining is one of the culprits, [making] up about one-third of the activity," Alvarez said. "More than half of the injection-type attacks are actually attributed to botnet-based activity launching command injection (CMDi) Local File Intrusion (LFI) attacks from hundreds of unique source addresses."
Best Practices for 2018
When it comes to the big cyber-security takeaways each year, Alvarez always expects to see more of the same, or the same with a new twist. It is the new twists, however, that can take enterprises by storm—for example, the destructive WannaCry and NotPetya ransomware attacks that propagated in 2017 coupled with sophisticated exploits.
"Despite using known flaws, it [WannaCry/NotPetya] spread like wildfire and led to billions in damage," Alvarez said. "We can hope that the massive, destructive attacks of 2017 served as a wake-up call for companies to put renewed efforts on many of the security basics that have been lacking."
One of the security basics that Alvarez sees lacking in many organizations is any type of real incident response plan. She said there are companies that still don't understand the distinction between a disaster recovery plan and a comprehensive incident response plan, which is needed to respond effectively and minimize damages.
"Incident response plans should also be practiced and tested on a regular basis," Alvarez recommends. "Have your retainers set up in advance for outside legal counsel, incident response and crisis communications."
Beyond having an incident response plan, IBM recommends that organizations identify what their potential targets are through risk assessment and security testing services. Alvarez said that everything from apps to internet of things (IoT) devices to even humans can undergo penetration testing. She added that a risk assessment of cloud deployments should also be conducted, particularly given the growing breaches as a result of human error in the cloud.
"Organizations should also keep in mind the security basics which will continue to remain priorities for the foreseeable future—backing up data, educating employees to avoid phishing attempts and maintaining regular patching routines," she said.
Sean Michael Kerner is a senior editor at eWEEK and InternetNews.com. Follow him on Twitter @TechJournalist.