AWS brings machine learning to cloud security, identifying threats automatically
Amazon Web Services (AWS) announced on Monday its new service Amazon Macie. The hosted security service uses machine learning to classify the sensitivity of customers' data in Amazon S3 and monitor and report any risks and unusual access, according to a corporate press release.
Macie recognizes sensitive data such as personal identity information or intellectual property and provides customers with dashboards and alerts that give visibility onto how this data is being accessed or moved, according to AWS.
Autodesk, Edmunds and Netflix are already testing Amazon Macie. The service is currently only for data stored in Amazon S3, but Amazon says support for additional AWS data stores will arrive later this year.
Organizations that use public cloud services like AWS or Microsoft Azure can find it more difficult to track their informational assets because not all potentially sensitive enterprise data is stored within the confines of the company’s data center.
If Macie works as promised, it could be a significant boon to cloud users concerned about sensitive data, providing them an early warning that their data is potentially unsafe.
It is worth mentioning that in recent months, Verizon and the WWE had two unknown vulnerabilities in their Amazon S3 buckets. With a tool like the new Macie service, the unintentionally open servers could have been flagged sooner. Several studies have found organizations often go several days or even months before they become aware of a data breach. By that time, damage is often already done.
Macie is another example of how machine learning can aid the enterprise by cutting through huge volumes of data to determine the most critical place IT staff need to focus their time.