From robotic meeting strains to self-driving vehicles, automated processes powered by synthetic intelligence (AI) are reshaping society in vital methods. But AI cannot do every little thing by itself — in reality, many organizations are starting to acknowledge that automation typically capabilities finest when it really works hand in hand with a human operator. Likewise, people can typically function extra effectively and successfully once they obtain a serving to hand from well-trained AI. Cybersecurity — notably id safety — is an ideal instance of a discipline the place augmenting the human contact with AI has produced extraordinarily optimistic outcomes.
Automation Is No Longer Optional
Consider the quantity of identities that exist in at the moment’s environments. Users, units, purposes, servers, cloud companies, databases, DevOps containers, and numerous different entities (each actual and digital) now have identities that must be managed. In addition, fashionable workers make the most of a variety of applied sciences and knowledge so as to be productive in enterprise environments. Together, these two dynamics create a problem for id safety — at at the moment’s scale, understanding which identities want entry to what programs has moved nicely past human capability.
This is essential as a result of cybercriminals are focusing on identities with elevated frequency. The most up-to-date “Verizon Data Breach Investigations Report” (DBIR) indicated that credential knowledge is now utilized in practically 50% of breaches, and stolen credentials are one of the widespread methods attackers are capable of compromise identities. Attackers use a wide range of strategies to acquire these credentials, however social engineering is maybe the most well-liked. People make errors, and attackers have gotten superb at figuring out methods to trick folks into making them. This is a serious a part of what makes at the moment’s attackers so troublesome to cease: Human beings are sometimes the weak level, and human beings can’t be patched. Designing a preventative answer that stops 100% of assaults merely is not doable.
Shifting the Focus to Containment
This is not to say that preventative measures like worker training, multifactor authentication, and frequent password modifications aren’t essential — they’re. But in addition they aren’t sufficient. Eventually, a decided attacker will discover a susceptible id to compromise, and the group might want to know what programs it had entry to and whether or not these privileges exceeded its precise wants. If an accountant has their consumer id compromised, that could be a downside — however it’s a downside that needs to be restricted to the accounting division. But in a company the place overprovisioning is widespread, an attacker who compromises a single id might need entry to any variety of programs. This downside is extra widespread than you would possibly assume — when a company has tens of hundreds of identities to handle, guaranteeing that every one has privileges that line up with its important capabilities is troublesome.
At least, it was once. Applied to id safety, AI-based applied sciences have made it doable to not solely assist enterprises handle id permissions at scale, however evolve id safety choices over time to make sure these choices match the shifting wants and dynamics of the enterprise. AI will be skilled to establish patterns that standard, human customers would by no means discover. For instance, they could search for permissions which might be hardly ever used and suggest that they be revoked — in any case, if they are not getting used, why threat permitting an attacker to use them? These instruments will be skilled to establish when entry to sure knowledge is often requested by the identical sort of consumer. They can then flag that data to an IT group member, who can choose whether or not extra permissions are warranted.
By figuring out these patterns, AI-based id instruments may also help to determine extra applicable permissions for identities throughout the group, whereas additionally offering IT workers with the data they should make knowledgeable choices as circumstances change. By eliminating extraneous, pointless permissions, AI instruments be certain that compromising a single id is not going to grant an attacker free reign all through your entire system. They additionally imply that, removed from impeding productiveness, the IT group can improve it. By shortly figuring out when it’s protected and applicable to grant extra permissions, they will be sure all identities underneath administration have entry to the know-how and knowledge they want, once they want it. None of this could be doable with out human beings and AI working hand in hand.
AI-Driven Identity Security Is the Future
Gone are the times when managing identities and their permissions could possibly be completed manually — at the moment, guaranteeing that every id has the fitting stage of entry can solely be completed with vital assist from synthetic intelligence-based know-how. By augmenting the human contact with AI, organizations can mix the velocity and accuracy of automation with the contextual judgment of human decision-making. Together, they may also help organizations extra successfully handle their identities and entitlements whereas considerably limiting the impression of any potential assault.
About the Author
Grady Summers has a wide range of know-how and management positions spanning over 20 years and now serves because the Executive Vice President of Product at SailPoint. Grady is chargeable for driving SailPoint’s know-how highway map and answer technique, guaranteeing robust and constant execution throughout SailPoint’s id portfolio.