SlideShare a Scribd company logo
1 of 19
Using a Cognitive Analytic Approach to Enhance
Cybersecurity on Oil and Gas OT Systems
Philippe Herve
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Communication with OT systems has traditionally been limited
▶ OT systems are old and outdated, and generally have a low degree of
connectivity between systems
▶ OT physical assets and control systems typically do not come with
adaptors for linking to IT networks
▶ The information used by one level of an OT system cannot be
understood by another
▶ OT design prioritizes safety, efficiency, and constant availability—not
confidentiality
▶ Volatile drilling environments and the need for constant production
can both make minor technical issues catastrophic
IoT is increasing connectivity of OT systems
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
The old paradigm for security is no longer scalable
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Evolving Landscape: With the number of connected devices and new threats growing exponentially
each year, machine scale is required to keep up with the evolving threat landscape
Number of connected devices
that will need to be secured in
2020, up from 9B in 2012
Cyber attacks originate from
the endpoint and propagate
through the network
Number of new malicious
threats created last year by
hackers around the globe,
up from 47M in 2010
Of connected devices
will be IoT
Ransomware payments were made
from corporations to hackers in 2015
325M
Expected increase in ransomware
payments in 2016
4x
The initial detection rate of newly
created malware by traditional
(signature based) antivirus solutions
<25%
The time it takes most traditional
antivirus vendors to detect a new
virus
4 weeks
50B
600M
50%
95%
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
The signature-based approach to endpoint protection is
broken. A new approach is needed to keep up with the
evolving threat landscape.
Traditional perimeter detection can no longer
keep up with the greater number of outside
connections in OT networks
Traditional endpoint detection can no longer
keep up with the proliferation of new sensors
and endpoints to guard
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Combine anomaly detection with machine learning for more robust
cyber protection
This is most effective when powered by machine learning.
OT systems are designed for repeatable communications. The
expected signals and behaviors of OT components are well defined,
making anomalies particularly uncommon—and particularly easy to
identify.
Anomaly detection is designed to monitor the behavior of
endpoint devices within the network and flag any unusual
behaviors or abnormal signals being sent out.
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Machine learning is inspired by how the human brain operates
Enables machines to
penetrate the
complexity of data to
identify associations
Presents powerful
techniques to handle
unstructured data
Continuously learns,
not only from previous
insights, but also for
new data entering the
system
Provides NLP support
to enable human to
machine and machine
to human
communication
Does not require
rules, instead relies on
hypothesis generation
built on analyzed data
Processes
information
Draws
conclusions
Codifies instincts &
Experiences into learning
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Subtler cyberattacks may involve unusual behaviors that still fall
within the normal rules of operation.
EX: A control system suddenly tells a device that regulates valves
to close a valve that is usually left open. This is a normal type of
message sent between the correct devices—but the content and
timing is statistically unusual.
Rules-based anomaly detection would not catch this, but a
machine learning system would.
Machine learning is inspired by how the human brain operates
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
How machine learning-powered anomaly detection could
have caught Stuxnet
A learning anomaly detection program could have caught Stuxnet at a variety of stages:
Operates outside PLCs, so would not have
been fooled by Stuxnet’s false signals from
PLCs that operations were normal and
would have detected anomalous
operations in centrifuges
Initial propagation of Stuxnet worm
between nodes in system would have
been flagged as anomalous before it could
reach the targeted programmable logic
controllers
Would have recognized that a worm had
found its way into the device and blocked
it from executing, therefore preventing it
from sending sabotaging orders to Iran’s
nuclear centrifuges
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Case study: Identifying anomalous
traffic with a learning algorithm
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
The problem
We were contracted by an industrial leader in gas, technologies, and services to investigate
their Intranet traffic in August of 2016.
Using a proprietary profile-based threat detection algorithm, a sample of the client company’s
firewall logs were examined.
Sample:
A single firewall
200,000
Cisco ASA
log lines
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Approach
The threat detection algorithm was trained on the firewall’s activity to build a profile of normal traffic
patterns. The log data given to the cognitive security company contained information as follows:
The algorithm then created a
profile of suspicious or
potentially malicious behavior
by studying the behavior
patterns of blocked traffic.
▶ Total Events – 200,001
▶ Blocked Events – 20,037
▶ Allowed Events – 105,000 (Connection Terminated)
▶ Other – 74,964 (Connection Created, Trans. Slot Deleted/Created, etc.)
▶ Distinct Client IP addresses in block vs. allow events – 945
▶ Most Common Client IP Address – 172.21.71.48 with 40,771 events
▶ Distinct Server IP addresses – 236
▶ Most Common Server IP Address – 172.21.66.37 with 40,363 events
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Methodology
The algorithm built its models by analyzing all inter-IP communication and building a feature set of each IP
address, including but not limited to the following created features:
▶Average time of access
▶Standard deviation of access time
▶Number of events
▶Number of distinct servers accessed
▶Number of client ports used
▶Percentage of server ports that were 80
▶Percentage of client ports that were
80
▶Percentage of time where server port
was different from client port
The dynamic model of suspicious behavior was then built off of these features using an automated model building
solution that included logistic regression, Bayesian tree-based models, support vector machines, and neural networks.
▶Number of distinct client hostnames
used
▶Number of distinct server ports used
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Results
The company’s OT network was not air gapped and was in
communication with outside entities. What resulted was an
algorithm that presented two potential conclusions about
any suspicious network activity:
traffic that should have been
blocked, but was not
traffic that was blocked,
but should not have been
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Results
▶ The model only agreed with 79%of the classifications
made by the client’s firewall.
▶ 1,749 IP addresses that the firewall monitored during
the time period of traffic under examination were false
negatives that should have been mitigated instead
▶ Only 29% of events associated with these suspicious IP
addresses were blocked
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Results
Example of threat caught by algorithm:
This turned out to be an
employee using VPN
against company
regulations.
Source: whois.arin.net
IP Address: 70.196.67.207 (United States)
Name: WIRELESSDATANETWORK
Handle: NET-70-192-0-0-1
Registration Date: 6/10/04
Range: 70.192.0.0-70.223.255.255
Org: Cellco Partnership DBA AT&T
Org Handle: CLLC
Address: 500 Jefferson Valley Drive
City: Bedminster
State/Province: NJ
Postal Code: 07039
Country: UNITED STATES
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Conclusion
IoT is changing the face of cybersecurity for OT in oil and gas. These changes in the structure of systems—as
well as the growing onslaught of new malware and zero-day attacks—require a change in the approach to
cybersecurity, both for perimeters and endpoints.
This case study demonstrates that while traditional security solutions can no longer protect OT systems,
machine learning solutions can.
As both our devices and our
threats become more intelligent,
so must our security systems.
OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
Acknowledgements / Thank You / Questions

More Related Content

What's hot

Security Analytics and Big Data: What You Need to Know
Security Analytics and Big Data: What You Need to KnowSecurity Analytics and Big Data: What You Need to Know
Security Analytics and Big Data: What You Need to KnowMapR Technologies
 
Power System Cybersecurity: Threats, Challenges, and Barriers
Power System Cybersecurity: Threats, Challenges, and Barriers Power System Cybersecurity: Threats, Challenges, and Barriers
Power System Cybersecurity: Threats, Challenges, and Barriers Nathan Wallace, PhD, PE
 
Power System Cybersecurity: Barriers and Challenges
Power System Cybersecurity: Barriers and Challenges Power System Cybersecurity: Barriers and Challenges
Power System Cybersecurity: Barriers and Challenges Nathan Wallace, PhD, PE
 
Operationalizing Security Intelligence
Operationalizing Security IntelligenceOperationalizing Security Intelligence
Operationalizing Security IntelligenceSplunk
 
IEEE PES GM 2017 Cybersecurity Panel Talk
IEEE PES GM 2017 Cybersecurity Panel TalkIEEE PES GM 2017 Cybersecurity Panel Talk
IEEE PES GM 2017 Cybersecurity Panel TalkNathan Wallace, PhD, PE
 
Manoj purandare - Stratergy towards an Effective Security Operations Centre -...
Manoj purandare - Stratergy towards an Effective Security Operations Centre -...Manoj purandare - Stratergy towards an Effective Security Operations Centre -...
Manoj purandare - Stratergy towards an Effective Security Operations Centre -...Manoj Purandare ☁
 
Cognitive Security: How Artificial Intelligence is Your New Best Friend
Cognitive Security: How Artificial Intelligence is Your New Best FriendCognitive Security: How Artificial Intelligence is Your New Best Friend
Cognitive Security: How Artificial Intelligence is Your New Best FriendSparkCognition
 
Stay out of headlines for non compliance or data breach
Stay out of headlines for non compliance or data breachStay out of headlines for non compliance or data breach
Stay out of headlines for non compliance or data breachSridhar Karnam
 
Best Practices for Scoping Infections and Disrupting Breaches
Best Practices for Scoping Infections and Disrupting BreachesBest Practices for Scoping Infections and Disrupting Breaches
Best Practices for Scoping Infections and Disrupting BreachesSplunk
 
Demystifying Security Analytics: Data, Methods, Use Cases
Demystifying Security Analytics: Data, Methods, Use CasesDemystifying Security Analytics: Data, Methods, Use Cases
Demystifying Security Analytics: Data, Methods, Use CasesPriyanka Aash
 
Building a Next-Generation Security Operations Center (SOC)
Building a Next-Generation Security Operations Center (SOC)Building a Next-Generation Security Operations Center (SOC)
Building a Next-Generation Security Operations Center (SOC)Sqrrl
 
Network Security‬ and Big ‪‎Data Analytics‬
Network Security‬ and Big ‪‎Data Analytics‬Network Security‬ and Big ‪‎Data Analytics‬
Network Security‬ and Big ‪‎Data Analytics‬Allot Communications
 
Steven Keil - BYODAWSCYW (Bring Your Own Device And Whatever Security Control...
Steven Keil - BYODAWSCYW (Bring Your Own Device And Whatever Security Control...Steven Keil - BYODAWSCYW (Bring Your Own Device And Whatever Security Control...
Steven Keil - BYODAWSCYW (Bring Your Own Device And Whatever Security Control...centralohioissa
 
Big Data Security Analytics (BDSA) with Randy Franklin
Big Data Security Analytics (BDSA) with Randy FranklinBig Data Security Analytics (BDSA) with Randy Franklin
Big Data Security Analytics (BDSA) with Randy FranklinSridhar Karnam
 
Using Data Science for Cybersecurity
Using Data Science for CybersecurityUsing Data Science for Cybersecurity
Using Data Science for CybersecurityVMware Tanzu
 
Big Data Analytics for Cyber Security: A Quick Overview
Big Data Analytics for Cyber Security: A Quick OverviewBig Data Analytics for Cyber Security: A Quick Overview
Big Data Analytics for Cyber Security: A Quick OverviewFemi Ashaye
 
SplunkLive! Customer Presentation - Dow Jones
SplunkLive! Customer Presentation - Dow JonesSplunkLive! Customer Presentation - Dow Jones
SplunkLive! Customer Presentation - Dow JonesSplunk
 
How is ai important to the future of cyber security
How is ai important to the future of cyber security How is ai important to the future of cyber security
How is ai important to the future of cyber security Robert Smith
 

What's hot (20)

Security Analytics and Big Data: What You Need to Know
Security Analytics and Big Data: What You Need to KnowSecurity Analytics and Big Data: What You Need to Know
Security Analytics and Big Data: What You Need to Know
 
Power System Cybersecurity: Threats, Challenges, and Barriers
Power System Cybersecurity: Threats, Challenges, and Barriers Power System Cybersecurity: Threats, Challenges, and Barriers
Power System Cybersecurity: Threats, Challenges, and Barriers
 
Power System Cybersecurity: Barriers and Challenges
Power System Cybersecurity: Barriers and Challenges Power System Cybersecurity: Barriers and Challenges
Power System Cybersecurity: Barriers and Challenges
 
Operationalizing Security Intelligence
Operationalizing Security IntelligenceOperationalizing Security Intelligence
Operationalizing Security Intelligence
 
IEEE PES GM 2017 Cybersecurity Panel Talk
IEEE PES GM 2017 Cybersecurity Panel TalkIEEE PES GM 2017 Cybersecurity Panel Talk
IEEE PES GM 2017 Cybersecurity Panel Talk
 
Manoj purandare - Stratergy towards an Effective Security Operations Centre -...
Manoj purandare - Stratergy towards an Effective Security Operations Centre -...Manoj purandare - Stratergy towards an Effective Security Operations Centre -...
Manoj purandare - Stratergy towards an Effective Security Operations Centre -...
 
Cognitive Security: How Artificial Intelligence is Your New Best Friend
Cognitive Security: How Artificial Intelligence is Your New Best FriendCognitive Security: How Artificial Intelligence is Your New Best Friend
Cognitive Security: How Artificial Intelligence is Your New Best Friend
 
A case for Managed Detection and Response
A case for Managed Detection and ResponseA case for Managed Detection and Response
A case for Managed Detection and Response
 
Stay out of headlines for non compliance or data breach
Stay out of headlines for non compliance or data breachStay out of headlines for non compliance or data breach
Stay out of headlines for non compliance or data breach
 
Best Practices for Scoping Infections and Disrupting Breaches
Best Practices for Scoping Infections and Disrupting BreachesBest Practices for Scoping Infections and Disrupting Breaches
Best Practices for Scoping Infections and Disrupting Breaches
 
Demystifying Security Analytics: Data, Methods, Use Cases
Demystifying Security Analytics: Data, Methods, Use CasesDemystifying Security Analytics: Data, Methods, Use Cases
Demystifying Security Analytics: Data, Methods, Use Cases
 
Building a Next-Generation Security Operations Center (SOC)
Building a Next-Generation Security Operations Center (SOC)Building a Next-Generation Security Operations Center (SOC)
Building a Next-Generation Security Operations Center (SOC)
 
Network Security‬ and Big ‪‎Data Analytics‬
Network Security‬ and Big ‪‎Data Analytics‬Network Security‬ and Big ‪‎Data Analytics‬
Network Security‬ and Big ‪‎Data Analytics‬
 
cb-EDR-V7_a4_Digital
cb-EDR-V7_a4_Digitalcb-EDR-V7_a4_Digital
cb-EDR-V7_a4_Digital
 
Steven Keil - BYODAWSCYW (Bring Your Own Device And Whatever Security Control...
Steven Keil - BYODAWSCYW (Bring Your Own Device And Whatever Security Control...Steven Keil - BYODAWSCYW (Bring Your Own Device And Whatever Security Control...
Steven Keil - BYODAWSCYW (Bring Your Own Device And Whatever Security Control...
 
Big Data Security Analytics (BDSA) with Randy Franklin
Big Data Security Analytics (BDSA) with Randy FranklinBig Data Security Analytics (BDSA) with Randy Franklin
Big Data Security Analytics (BDSA) with Randy Franklin
 
Using Data Science for Cybersecurity
Using Data Science for CybersecurityUsing Data Science for Cybersecurity
Using Data Science for Cybersecurity
 
Big Data Analytics for Cyber Security: A Quick Overview
Big Data Analytics for Cyber Security: A Quick OverviewBig Data Analytics for Cyber Security: A Quick Overview
Big Data Analytics for Cyber Security: A Quick Overview
 
SplunkLive! Customer Presentation - Dow Jones
SplunkLive! Customer Presentation - Dow JonesSplunkLive! Customer Presentation - Dow Jones
SplunkLive! Customer Presentation - Dow Jones
 
How is ai important to the future of cyber security
How is ai important to the future of cyber security How is ai important to the future of cyber security
How is ai important to the future of cyber security
 

Similar to Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems

Standards based security for energy utilities
Standards based security for energy utilitiesStandards based security for energy utilities
Standards based security for energy utilitiesNirmal Thaliyil
 
Light sec for utilities and critical infrastructure white paper
Light sec for utilities and critical infrastructure white paperLight sec for utilities and critical infrastructure white paper
Light sec for utilities and critical infrastructure white paperGeorge Wainblat
 
Ciss previsionnotes
Ciss previsionnotesCiss previsionnotes
Ciss previsionnotesmadunix
 
Mobile fraud detection using neural networks
Mobile fraud detection using neural networksMobile fraud detection using neural networks
Mobile fraud detection using neural networksVidhya Moorthy
 
White paper scada (2)
White paper scada (2)White paper scada (2)
White paper scada (2)Ivan Carmona
 
Deploying Network Taps for Improved Security
Deploying Network Taps for Improved SecurityDeploying Network Taps for Improved Security
Deploying Network Taps for Improved SecurityDatacomsystemsinc
 
Scada security presentation by Stephen Miller
Scada security presentation by Stephen MillerScada security presentation by Stephen Miller
Scada security presentation by Stephen MillerAVEVA
 
IRJET-Secured Approach for Authentication of Messages in Wireless Sensor Netw...
IRJET-Secured Approach for Authentication of Messages in Wireless Sensor Netw...IRJET-Secured Approach for Authentication of Messages in Wireless Sensor Netw...
IRJET-Secured Approach for Authentication of Messages in Wireless Sensor Netw...IRJET Journal
 
All Hope is Not Lost Network Forensics Exposes Today's Advanced Security Thr...
All Hope is Not LostNetwork Forensics Exposes Today's Advanced Security Thr...All Hope is Not LostNetwork Forensics Exposes Today's Advanced Security Thr...
All Hope is Not Lost Network Forensics Exposes Today's Advanced Security Thr...Savvius, Inc
 
Week 09_Cyber security u.pdf
Week 09_Cyber security u.pdfWeek 09_Cyber security u.pdf
Week 09_Cyber security u.pdfdhanywahyudi17
 
IRJET- A Review on Application of Data Mining Techniques for Intrusion De...
IRJET-  	  A Review on Application of Data Mining Techniques for Intrusion De...IRJET-  	  A Review on Application of Data Mining Techniques for Intrusion De...
IRJET- A Review on Application of Data Mining Techniques for Intrusion De...IRJET Journal
 
Analysis of exposed ICS//SCADA/IoT systems in Europe
Analysis of exposed ICS//SCADA/IoT systems in EuropeAnalysis of exposed ICS//SCADA/IoT systems in Europe
Analysis of exposed ICS//SCADA/IoT systems in EuropeFrancesco Faenzi
 
NetSpi Whitepaper: Hardening Critical Systems At Electrical Utilities
NetSpi Whitepaper: Hardening Critical Systems At Electrical UtilitiesNetSpi Whitepaper: Hardening Critical Systems At Electrical Utilities
NetSpi Whitepaper: Hardening Critical Systems At Electrical UtilitiesCoreTrace Corporation
 
IRJET- An Intrusion Detection Framework based on Binary Classifiers Optimized...
IRJET- An Intrusion Detection Framework based on Binary Classifiers Optimized...IRJET- An Intrusion Detection Framework based on Binary Classifiers Optimized...
IRJET- An Intrusion Detection Framework based on Binary Classifiers Optimized...IRJET Journal
 
Cybercrime future perspectives
Cybercrime future perspectivesCybercrime future perspectives
Cybercrime future perspectivesSensePost
 
CNIT 152: 9 Network Evidence
CNIT 152: 9 Network EvidenceCNIT 152: 9 Network Evidence
CNIT 152: 9 Network EvidenceSam Bowne
 
Critical Information Infrastructure Systems Worldwide
Critical Information Infrastructure Systems WorldwideCritical Information Infrastructure Systems Worldwide
Critical Information Infrastructure Systems WorldwideAngela Hays
 
Dr Dev Kambhampati | Electric Utilities Situational Awareness
Dr Dev Kambhampati | Electric Utilities Situational AwarenessDr Dev Kambhampati | Electric Utilities Situational Awareness
Dr Dev Kambhampati | Electric Utilities Situational AwarenessDr Dev Kambhampati
 

Similar to Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems (20)

Standards based security for energy utilities
Standards based security for energy utilitiesStandards based security for energy utilities
Standards based security for energy utilities
 
Light sec for utilities and critical infrastructure white paper
Light sec for utilities and critical infrastructure white paperLight sec for utilities and critical infrastructure white paper
Light sec for utilities and critical infrastructure white paper
 
Ciss previsionnotes
Ciss previsionnotesCiss previsionnotes
Ciss previsionnotes
 
Mobile fraud detection using neural networks
Mobile fraud detection using neural networksMobile fraud detection using neural networks
Mobile fraud detection using neural networks
 
White paper scada (2)
White paper scada (2)White paper scada (2)
White paper scada (2)
 
Deploying Network Taps for Improved Security
Deploying Network Taps for Improved SecurityDeploying Network Taps for Improved Security
Deploying Network Taps for Improved Security
 
Scada security presentation by Stephen Miller
Scada security presentation by Stephen MillerScada security presentation by Stephen Miller
Scada security presentation by Stephen Miller
 
IRJET-Secured Approach for Authentication of Messages in Wireless Sensor Netw...
IRJET-Secured Approach for Authentication of Messages in Wireless Sensor Netw...IRJET-Secured Approach for Authentication of Messages in Wireless Sensor Netw...
IRJET-Secured Approach for Authentication of Messages in Wireless Sensor Netw...
 
All Hope is Not Lost Network Forensics Exposes Today's Advanced Security Thr...
All Hope is Not LostNetwork Forensics Exposes Today's Advanced Security Thr...All Hope is Not LostNetwork Forensics Exposes Today's Advanced Security Thr...
All Hope is Not Lost Network Forensics Exposes Today's Advanced Security Thr...
 
Week 09_Cyber security u.pdf
Week 09_Cyber security u.pdfWeek 09_Cyber security u.pdf
Week 09_Cyber security u.pdf
 
Webinar: SecurePlanHealth Updates
Webinar: SecurePlanHealth UpdatesWebinar: SecurePlanHealth Updates
Webinar: SecurePlanHealth Updates
 
IRJET- A Review on Application of Data Mining Techniques for Intrusion De...
IRJET-  	  A Review on Application of Data Mining Techniques for Intrusion De...IRJET-  	  A Review on Application of Data Mining Techniques for Intrusion De...
IRJET- A Review on Application of Data Mining Techniques for Intrusion De...
 
Analysis of exposed ICS//SCADA/IoT systems in Europe
Analysis of exposed ICS//SCADA/IoT systems in EuropeAnalysis of exposed ICS//SCADA/IoT systems in Europe
Analysis of exposed ICS//SCADA/IoT systems in Europe
 
NetSpi Whitepaper: Hardening Critical Systems At Electrical Utilities
NetSpi Whitepaper: Hardening Critical Systems At Electrical UtilitiesNetSpi Whitepaper: Hardening Critical Systems At Electrical Utilities
NetSpi Whitepaper: Hardening Critical Systems At Electrical Utilities
 
IRJET- An Intrusion Detection Framework based on Binary Classifiers Optimized...
IRJET- An Intrusion Detection Framework based on Binary Classifiers Optimized...IRJET- An Intrusion Detection Framework based on Binary Classifiers Optimized...
IRJET- An Intrusion Detection Framework based on Binary Classifiers Optimized...
 
jhon ibrahim.ppt
jhon ibrahim.pptjhon ibrahim.ppt
jhon ibrahim.ppt
 
Cybercrime future perspectives
Cybercrime future perspectivesCybercrime future perspectives
Cybercrime future perspectives
 
CNIT 152: 9 Network Evidence
CNIT 152: 9 Network EvidenceCNIT 152: 9 Network Evidence
CNIT 152: 9 Network Evidence
 
Critical Information Infrastructure Systems Worldwide
Critical Information Infrastructure Systems WorldwideCritical Information Infrastructure Systems Worldwide
Critical Information Infrastructure Systems Worldwide
 
Dr Dev Kambhampati | Electric Utilities Situational Awareness
Dr Dev Kambhampati | Electric Utilities Situational AwarenessDr Dev Kambhampati | Electric Utilities Situational Awareness
Dr Dev Kambhampati | Electric Utilities Situational Awareness
 

More from SparkCognition

How to Use Artificial Intelligence to Minimize your Cybersecurity Attack Surface
How to Use Artificial Intelligence to Minimize your Cybersecurity Attack SurfaceHow to Use Artificial Intelligence to Minimize your Cybersecurity Attack Surface
How to Use Artificial Intelligence to Minimize your Cybersecurity Attack SurfaceSparkCognition
 
Cognitive Analysis With SparkSecure
Cognitive Analysis With SparkSecureCognitive Analysis With SparkSecure
Cognitive Analysis With SparkSecureSparkCognition
 
AWEA Cognitive Analytics for Predictive Futures
AWEA Cognitive Analytics for Predictive FuturesAWEA Cognitive Analytics for Predictive Futures
AWEA Cognitive Analytics for Predictive FuturesSparkCognition
 
Cyberattacks on the Rise Infographic
Cyberattacks on the Rise InfographicCyberattacks on the Rise Infographic
Cyberattacks on the Rise InfographicSparkCognition
 
Artificial Intelligence Application in Oil and Gas
Artificial Intelligence Application in Oil and GasArtificial Intelligence Application in Oil and Gas
Artificial Intelligence Application in Oil and GasSparkCognition
 

More from SparkCognition (6)

How to Use Artificial Intelligence to Minimize your Cybersecurity Attack Surface
How to Use Artificial Intelligence to Minimize your Cybersecurity Attack SurfaceHow to Use Artificial Intelligence to Minimize your Cybersecurity Attack Surface
How to Use Artificial Intelligence to Minimize your Cybersecurity Attack Surface
 
Cognitive Analysis With SparkSecure
Cognitive Analysis With SparkSecureCognitive Analysis With SparkSecure
Cognitive Analysis With SparkSecure
 
Ai in Cyber Warfare
Ai in Cyber WarfareAi in Cyber Warfare
Ai in Cyber Warfare
 
AWEA Cognitive Analytics for Predictive Futures
AWEA Cognitive Analytics for Predictive FuturesAWEA Cognitive Analytics for Predictive Futures
AWEA Cognitive Analytics for Predictive Futures
 
Cyberattacks on the Rise Infographic
Cyberattacks on the Rise InfographicCyberattacks on the Rise Infographic
Cyberattacks on the Rise Infographic
 
Artificial Intelligence Application in Oil and Gas
Artificial Intelligence Application in Oil and GasArtificial Intelligence Application in Oil and Gas
Artificial Intelligence Application in Oil and Gas
 

Recently uploaded

Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyAlfredo García Lavilla
 
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
"Federated learning: out of reach no matter how close",Oleksandr LapshynFwdays
 
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Mark Simos
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Patryk Bandurski
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Wonjun Hwang
 
Vector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesVector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesZilliz
 
Powerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time ClashPowerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time Clashcharlottematthew16
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):comworks
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupFlorian Wilhelm
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii SoldatenkoFwdays
 
Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Manik S Magar
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embeddingZilliz
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...Fwdays
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024Stephanie Beckett
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLScyllaDB
 
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxNavinnSomaal
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Mattias Andersson
 

Recently uploaded (20)

Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easy
 
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
"Federated learning: out of reach no matter how close",Oleksandr Lapshyn
 
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
 
Vector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesVector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector Databases
 
Powerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time ClashPowerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time Clash
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project Setup
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko
 
Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embedding
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQL
 
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptx
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?
 

Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems

  • 1. Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems Philippe Herve
  • 2. OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve Communication with OT systems has traditionally been limited ▶ OT systems are old and outdated, and generally have a low degree of connectivity between systems ▶ OT physical assets and control systems typically do not come with adaptors for linking to IT networks ▶ The information used by one level of an OT system cannot be understood by another ▶ OT design prioritizes safety, efficiency, and constant availability—not confidentiality ▶ Volatile drilling environments and the need for constant production can both make minor technical issues catastrophic
  • 3. IoT is increasing connectivity of OT systems OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 4. The old paradigm for security is no longer scalable OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 5. Evolving Landscape: With the number of connected devices and new threats growing exponentially each year, machine scale is required to keep up with the evolving threat landscape Number of connected devices that will need to be secured in 2020, up from 9B in 2012 Cyber attacks originate from the endpoint and propagate through the network Number of new malicious threats created last year by hackers around the globe, up from 47M in 2010 Of connected devices will be IoT Ransomware payments were made from corporations to hackers in 2015 325M Expected increase in ransomware payments in 2016 4x The initial detection rate of newly created malware by traditional (signature based) antivirus solutions <25% The time it takes most traditional antivirus vendors to detect a new virus 4 weeks 50B 600M 50% 95% OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 6. The signature-based approach to endpoint protection is broken. A new approach is needed to keep up with the evolving threat landscape. Traditional perimeter detection can no longer keep up with the greater number of outside connections in OT networks Traditional endpoint detection can no longer keep up with the proliferation of new sensors and endpoints to guard OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 7. Combine anomaly detection with machine learning for more robust cyber protection This is most effective when powered by machine learning. OT systems are designed for repeatable communications. The expected signals and behaviors of OT components are well defined, making anomalies particularly uncommon—and particularly easy to identify. Anomaly detection is designed to monitor the behavior of endpoint devices within the network and flag any unusual behaviors or abnormal signals being sent out. OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 8. Machine learning is inspired by how the human brain operates Enables machines to penetrate the complexity of data to identify associations Presents powerful techniques to handle unstructured data Continuously learns, not only from previous insights, but also for new data entering the system Provides NLP support to enable human to machine and machine to human communication Does not require rules, instead relies on hypothesis generation built on analyzed data Processes information Draws conclusions Codifies instincts & Experiences into learning OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 9. Subtler cyberattacks may involve unusual behaviors that still fall within the normal rules of operation. EX: A control system suddenly tells a device that regulates valves to close a valve that is usually left open. This is a normal type of message sent between the correct devices—but the content and timing is statistically unusual. Rules-based anomaly detection would not catch this, but a machine learning system would. Machine learning is inspired by how the human brain operates OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 10. How machine learning-powered anomaly detection could have caught Stuxnet A learning anomaly detection program could have caught Stuxnet at a variety of stages: Operates outside PLCs, so would not have been fooled by Stuxnet’s false signals from PLCs that operations were normal and would have detected anomalous operations in centrifuges Initial propagation of Stuxnet worm between nodes in system would have been flagged as anomalous before it could reach the targeted programmable logic controllers Would have recognized that a worm had found its way into the device and blocked it from executing, therefore preventing it from sending sabotaging orders to Iran’s nuclear centrifuges OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 11. Case study: Identifying anomalous traffic with a learning algorithm OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 12. The problem We were contracted by an industrial leader in gas, technologies, and services to investigate their Intranet traffic in August of 2016. Using a proprietary profile-based threat detection algorithm, a sample of the client company’s firewall logs were examined. Sample: A single firewall 200,000 Cisco ASA log lines OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 13. Approach The threat detection algorithm was trained on the firewall’s activity to build a profile of normal traffic patterns. The log data given to the cognitive security company contained information as follows: The algorithm then created a profile of suspicious or potentially malicious behavior by studying the behavior patterns of blocked traffic. ▶ Total Events – 200,001 ▶ Blocked Events – 20,037 ▶ Allowed Events – 105,000 (Connection Terminated) ▶ Other – 74,964 (Connection Created, Trans. Slot Deleted/Created, etc.) ▶ Distinct Client IP addresses in block vs. allow events – 945 ▶ Most Common Client IP Address – 172.21.71.48 with 40,771 events ▶ Distinct Server IP addresses – 236 ▶ Most Common Server IP Address – 172.21.66.37 with 40,363 events OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 14. Methodology The algorithm built its models by analyzing all inter-IP communication and building a feature set of each IP address, including but not limited to the following created features: ▶Average time of access ▶Standard deviation of access time ▶Number of events ▶Number of distinct servers accessed ▶Number of client ports used ▶Percentage of server ports that were 80 ▶Percentage of client ports that were 80 ▶Percentage of time where server port was different from client port The dynamic model of suspicious behavior was then built off of these features using an automated model building solution that included logistic regression, Bayesian tree-based models, support vector machines, and neural networks. ▶Number of distinct client hostnames used ▶Number of distinct server ports used OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 15. Results The company’s OT network was not air gapped and was in communication with outside entities. What resulted was an algorithm that presented two potential conclusions about any suspicious network activity: traffic that should have been blocked, but was not traffic that was blocked, but should not have been OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 16. Results ▶ The model only agreed with 79%of the classifications made by the client’s firewall. ▶ 1,749 IP addresses that the firewall monitored during the time period of traffic under examination were false negatives that should have been mitigated instead ▶ Only 29% of events associated with these suspicious IP addresses were blocked OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 17. Results Example of threat caught by algorithm: This turned out to be an employee using VPN against company regulations. Source: whois.arin.net IP Address: 70.196.67.207 (United States) Name: WIRELESSDATANETWORK Handle: NET-70-192-0-0-1 Registration Date: 6/10/04 Range: 70.192.0.0-70.223.255.255 Org: Cellco Partnership DBA AT&T Org Handle: CLLC Address: 500 Jefferson Valley Drive City: Bedminster State/Province: NJ Postal Code: 07039 Country: UNITED STATES OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 18. Conclusion IoT is changing the face of cybersecurity for OT in oil and gas. These changes in the structure of systems—as well as the growing onslaught of new malware and zero-day attacks—require a change in the approach to cybersecurity, both for perimeters and endpoints. This case study demonstrates that while traditional security solutions can no longer protect OT systems, machine learning solutions can. As both our devices and our threats become more intelligent, so must our security systems. OTC-27895-MS • Using a Cognitive Analytic Approach to Enhance Cybersecurity on Oil and Gas OT Systems • Herve
  • 19. Acknowledgements / Thank You / Questions