Why companies should be using AI to fight cyberattacks

1 year ago 158

Cyberattackers usage AI, truthful wherefore not use it arsenic a defense? One adept explains wherefore AI tin instrumentality your cybersecurity to the adjacent level of protection.


Image: iStockPhoto/maxkabakov

In immoderate debate, determination are ever astatine slightest 2 sides. That reasoning besides applies to whether oregon not it is simply a bully thought to usage artificial intelligence exertion to effort stemming the advantages of cybercriminals who are already utilizing AI to amended their occurrence ratio. 

SEE: Google Chrome: Security and UI tips you request to know  (TechRepublic Premium)

In an email exchange, I asked Ramprakash Ramamoorthy, manager of probe astatine ManageEngine, a part of Zoho Corporation, for his thoughts connected the matter. Ramamoorthy is firmly connected the affirmative broadside for utilizing AI to combat cybercrime. He said, "The lone mode to combat cybercriminals utilizing AI-enhanced attacks is to combat occurrence with occurrence and employment AI countermeasures."

Why take AI successful cybersecurity?

An evident question is: Why adhd different costly exertion to a company's cybersecurity platform, particularly successful a section that galore precocious absorption types see to person a unspeakable instrumentality connected investment? Ramamoorthy offered the pursuing reasons:

  • Enterprise information and privateness practices person go the practice of the trustworthiness of a business. A information breach oregon escaped privateness practices mightiness harm an organization's estimation to the grade that it could thrust distant customers to competitors, irrespective of the competitiveness of your offering.
  • It's lone just that you enactment your champion ft guardant to marque definite you enactment connected apical of the cybersecurity game. Deploying evolving technologies similar AI into your information practices tin nonstop beardown signals to your customers that you person been taking them precise seriously, and you're successful it for the agelong term.

Besides maintaining a bully nationalist image, Ramamoorthy said helium believes AI tin assistance an enactment enactment up of cyberattackers. We each cognize the pandemic satellite has democratized entree to delicate data. Confidential accusation is nary longer restricted to backstage networks oregon firm devices but tin beryllium accessed from anyplace connected immoderate device. 

"This gives hackers aggregate imaginable entree points to entree your confidential endeavor information illegally," Ramamoorthy said. "Attackers usage almighty techniques similar AI to exploit unsuspecting end-users to summation entree to privileged accusation by compromising said entree points."

SEE: Password breach: Why popular civilization and passwords don't premix (free PDF) (TechRepublic)

Another disadvantage is that accepted (non-AI) information approaches person ever worked based connected static thresholds. Attackers tin crippled the strategy by flying nether the radar of static thresholds.

With that successful mind, Ramamoorthy past asked wherefore organizations aren't utilizing the aforesaid exertion to combat back? The clip is ripe for upping the information and privateness extortion crippled with the assistance of AI. Ramamoorthy offered respective real-world cyberattack scenarios and however AI would assistance cybercrime-fighters.

  1. Example: An enactment with a SIEM solution has it acceptable to alert erstwhile the fig of failed logins to entree proprietary accusation reaches 10 per minute. A brute-forcing attacker tin inactive bash 9 failed logins per infinitesimal and locomotion distant unidentified.
    Solution: Set elastic thresholds with minimal-to-no quality intervention. Also, AI tin show login patterns and acceptable up thresholds depending connected aggregate variables similar clip of day, time of the week, and different caller trends successful accusation access. For example, a Monday greeting astatine 9 AM and a Saturday greeting astatine 3 AM mightiness request antithetic thresholds.
  2. Example: An ill-configured threshold could pb to alert fatigue to whomever is liable for monitoring SIEM strategy alerts. 
    Solution: AI tin mitigate alert fatigue by identifying frequent, rare, unseen patterns and mounting the alert precedence accordingly.
  3. Example: It is astir intolerable for cybersecurity unit to show entree to each imaginable ransomware and phishing website. 
    Solution: AI tin beryllium deployed astatine endpoints to assistance place and quarantine malicious websites, thereby enabling amended data-access practices combined with techniques similar multifactor authentication and zero-trust security.

Can AI amended information of information stored successful the cloud?

Ramamoorthy said helium believes AI tin guarantee amended information crossed the tech stack—from unreality deployments to endpoints accessing data. "Rule-based systems mightiness not beryllium capable to drawback information vulnerabilities crossed the stack and mightiness request analyzable rules to beryllium written and maintained implicit time," Ramamoorthy said. "With AI, the thresholds are automatically acceptable depending connected the inclination and seasonal patterns successful the data."

He continued, "At the unreality level, AI tin bounds entree to privileged accusation and debar assorted attacks similar Distributed Denial of Services, zero-day exploits, etc."

What to look for successful AI-security solutions

According to Ramamoorthy, it is important to guarantee the selected AI solution envelopes successful the full stack. Also, SIEM products with AI-based UEBA (User and Entity Behavior Analysis) tools would assistance guarantee the information of captious systems.

He besides noted endpoint-protection products are starting to see AI-based features specified arsenic ransomware recognition and malware mitigation.

Deploy AI capabilities sooner alternatively than later

Ramamoorthy suggested utilizing AI successful cybersecurity is an fantabulous mode to debar being the lowest-hanging effect connected the integer tree, arsenic not galore organizations are present employing AI cybersecurity solutions. That is not existent with cybercriminals; they're keen connected AI and deploying much AI-enhanced cyberattack technology each day.

There is simply a crushed Ramamoorthy utilized the examples helium did. He explained wherefore successful his parting comments: "Embracing AI-based UEBA modules arsenic portion of an organization's SIEM solution should beryllium the archetypal step, arsenic it is simply a adjuvant mode of monitoring users and entities, arsenic good arsenic identifying suspicious patterns aboriginal on."

Data, Analytics and AI Newsletter

Learn the latest quality and champion practices astir information science, large information analytics, and artificial intelligence. Delivered Mondays

Sign up today

Also see

Read Entire Article