pogonici - Fotolia
A house needs a strong foundation to build a new addition. Similarly, a company needs a strong foundation to build upon content security. Without a solid framework, details may slip through the cracks and increase the possibility of problems down the road.
AI tools can strengthen enterprise security, but they won't reach their full potential unless a structured framework is in place.
AI can enhance content security by monitoring data flows and recognizing anomalies. But the underlying framework is primarily an operational concern, beginning with how networks and systems are managed both on premises and in the cloud. AI tools and technologies can then manage the repetitive and time-consuming tasks of continuous monitoring and sense potential threats.
Before incorporating AI into content security, companies should focus on their operational activities -- specifically policies, procedures and operational-level business rules. It is essential to invest the time and effort to describe these operational activities, as well as define how content should securely flow across an extended enterprise. A security audit can assess capabilities of existing system- and network-level services. Consider, for example, how unsecured endpoints such as mobile devices and web browsers pose risks to end-to-end content flows.
When it comes to content security, there are multiple legal, compliance, risk management and business concerns. For instance, a business might restrict drafts of a proposal to members of the bid team, and once the team captain approves them, they can be made available to other company leaders. Organizations rely on business processes to ensure the availability, usability, consistency, integrity and security of the information they collect and maintain to run their businesses. It's important to categorize business content in a systematic manner based on their sensitivity and business risk, and to have operational processes in place to keep these classifications current.
Within data-savvy organizations, data stewards have well-established operational roles. They ensure that all employees follow data governance processes and also enforce guidelines when managing structured data within databases. Content stewards should have comparable roles enforcing procedures related to documents, email messages, digital assets and other kinds of digitized information that businesses manage within various types of content repositories. Content stewards should be able to track end-to-end content flows for sensitive business documents and describe the risks that innovative security technologies can address.
Signals and patterns
Signals and patterns also matter for content security. It's no longer enough for organizations to rely on identity management, authorization and access controls -- conventional ways of securing content managed within shared repositories. The spread of connected devices and processes within an enterprise adds to security risks, which businesses can mitigate through using AI in content security.
Suppose a company is monitoring document downloads from a shared repository. Threat hunters assess risks to events in the cloud while classifiers automatically recognize content that requires protection. If the person or process requesting a download is in a remote location where the organization does not have a business presence, then something may be amiss. A content security tool such as Box Shield, which relies on machine learning to track patterns, can detect the anomaly and take predefined actions -- perhaps automatically shutting down the access port or notifying an administrator.
Microsoft also tackles these kinds of distributed risks within its cloud services. In particular, Azure Sentinel includes enhanced threat hunting capabilities for detecting suspicious network events, such as logins from anomalous IP addresses. The compliance center within Microsoft 365 can track content flowing through SharePoint and Outlook repositories by sensitivity level and uses machine learning to train classifiers by categories that are unique to an organization. Not surprisingly, Microsoft demonstrates how a cloud-powered repository provides the foundation for content security and governance across the extended enterprise.
How AI makes a difference
It is still the early days for applying AI to content security. AI tools -- powered by machine learning, natural language processing and even image recognition algorithms -- can ensure new levels of content security. Businesses can expect these inferencing capabilities to become more intelligent with the addition of advanced algorithms. But first, companies need to define the dots and map how they relate to content security. They need to ensure that operational activities -- including people and processes -- reinforce enabling security technologies.
Content security relies on smart categories. When developing an enterprise information architecture, be sure to include metadata for governance and content security. AI tools enable businesses to more precisely monitor embedded signals about the security of content flows.