ZoneFox Augmented Intelligence (A.I.)

Similar documents
FOREST PRODUCTS: THE SHIFT TO DIGITAL ACCELERATES

Our Corporate Strategy Digital

The A.I. Revolution Begins With Augmented Intelligence. White Paper January 2018

2. The Crypto Story So Far

MORE POWER TO THE ENERGY AND UTILITIES BUSINESS, FROM AI.

technologies, Gigaom provides deep insight on the disruptive companies, people and technologies shaping the future for all of us.

PURPOSE OF THIS EBOOK

Human vs Computer. Reliability & Competition

THE USE OF ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING IN SPEECH RECOGNITION. A CS Approach By Uniphore Software Systems

How do you teach AI the value of trust?

THE AI REVOLUTION. How Artificial Intelligence is Redefining Marketing Automation

Copyright: Conference website: Date deposited:

Powering Human Capability

Development and Integration of Artificial Intelligence Technologies for Innovation Acceleration

2017/18 KEYNOTE OVERVIEW DIGITAL EVANGELIST PATTERN HUNTER TREND SPOTTER MEDIA COMMENTATOR STORY TELLER

Expression Of Interest

Ethics and technology

Personalised for: Rachel Parkes

The five senses of Artificial Intelligence

Looking ahead : Technology trends driving business innovation.

CAUTIOUS OPTIMISM MARKS THE ADOPTION OF AI AT PROXIMUS

Why we need to know what AI is. Overview. Artificial Intelligence is it finally arriving?

Transer Learning : Super Intelligence

Component Based Mechatronics Modelling Methodology

Author s Name Name of the Paper Session. DYNAMIC POSITIONING CONFERENCE October 10-11, 2017 SENSORS SESSION. Sensing Autonomy.

Find and analyse the most relevant patents for your research

Assessing and Integrating Emerging Technologies

The future of work. Artificial Intelligence series

The five senses of Artificial Intelligence. Why humanizing automation is crucial to the transformation of your business

The Five Senses of Intelligent Automation

Source: REUTERS/Reinhard Krause

Targeting a Safer World. Public Safety & Security

2018 Avanade Inc. All Rights Reserved.

SEPTEMBER, 2018 PREDICTIVE MAINTENANCE SOLUTIONS

Stepping up and Stepping out

The Implementation of Artificial Intelligence and Machine Learning in a Computerized Chess Program

THE PRESENT AND THE FUTURE OF igaming

COMP219: Artificial Intelligence. Lecture 2: AI Problems and Applications

V. Adamchik Data Structures. Game Trees. Lecture 1. Apr. 05, Plan: 1. Introduction. 2. Game of NIM. 3. Minimax

How AI and wearables will take health to the next level - AI Med

AI & Law. What is AI?

Years 9 and 10 standard elaborations Australian Curriculum: Digital Technologies

Computer Log Anomaly Detection Using Frequent Episodes

AI for Autonomous Ships Challenges in Design and Validation

The Future of Patient Data The Global View Key Insights Berlin 18 April The world s leading open foresight program

BI TRENDS FOR Data De-silofication: The Secret to Success in the Analytics Economy

Notes from a seminar on "Tackling Public Sector Fraud" presented jointly by the UK NAO and H M Treasury in London, England in February 1998.

Machine Learning has been used in the real estate industry much longer than headlines and pitch decks suggest

Prof. Subramanian Ramamoorthy. The University of Edinburgh, Reader at the School of Informatics

Outline. Game Playing. Game Problems. Game Problems. Types of games Playing a perfect game. Playing an imperfect game

Making intelligent energy happen

Humanification Go Digital, Stay Human

Executive Summary Industry s Responsibility in Promoting Responsible Development and Use:

Expectations for Intelligent Computing

Operational Intelligence to Deliver Smart Solutions. Copyright 2015 OSIsoft, LLC

The Transformative Power of Technology

THE ACCOUNTANT S GUIDE TO AI STANDING ON THE SHOULDERS OF ROBOTS

in the New Zealand Curriculum

Point of View. Establishing a Culture of Digital Change within Universities

Tuning-CALOHEE Assessment Frameworks for the Subject Area of CIVIL ENGINEERING The Tuning-CALOHEE Assessment Frameworks for Civil Engineering offers

Industry 4.0: the new challenge for the Italian textile machinery industry

AGENTS AND AGREEMENT TECHNOLOGIES: THE NEXT GENERATION OF DISTRIBUTED SYSTEMS

Un nuevo origen 7 tendencias claves. A New Origin: 7 Key Trends

Digital Government Experience Centre. Accelerate your digital transformation

Foundations of AI. 6. Adversarial Search. Search Strategies for Games, Games with Chance, State of the Art. Wolfram Burgard & Bernhard Nebel

By Mark Hindsbo Vice President and General Manager, ANSYS

UN-GGIM Future Trends in Geospatial Information Management 1

COMPREHENSIVE COMPETITIVE INTELLIGENCE MONITORING IN REAL TIME

Vorwerk Thermomix C O N S U L T A N C Y C A S E S T U D Y

Forward Looking Information

AI and machine learning get us one step closer to relevance at scale

AI powering Corporate Communications

Executive Summary. The process. Intended use

Digital Transformation Delivering Business Outcomes

OECD WORK ON ARTIFICIAL INTELLIGENCE

Keywords: Multi-robot adversarial environments, real-time autonomous robots

Artificial Intelligence for Games

Digital Transformation Delivering Business Outcomes

Global Standards Symposium. Security, privacy and trust in standardisation. ICDPPC Chair John Edwards. 24 October 2016

VIEW POINT CHANGING THE BUSINESS LANDSCAPE WITH COGNITIVE SERVICES

Louis Spaninks. National coordinator Human Capital Agenda ICT. dcypher Symposium 2017 Oct 4th Media Plaza Utrecht connects cybersecurity knowledge

How Explainability is Driving the Future of Artificial Intelligence. A Kyndi White Paper

OFFensive Swarm-Enabled Tactics (OFFSET)

Become digitally disruptive: The challenge to unlearn

Appendices master s degree programme Artificial Intelligence

C&D Summit 2018 / CIMON / May 31, 2018 / 2018 IBM Corporation. Presentation should start with this video:

Big Data & AI Governance: The Laws and Ethics

Situation Awareness in Network Based Command & Control Systems

DOW IMPROVES INSTRUMENT RELIABILITY 66% AND SAVES MILLIONS OF DOLLARS WITH REAL-TIME HART TECHNOLOGY

Our position. ICDPPC declaration on ethics and data protection in artificial intelligence

CPS331 Lecture: Intelligent Agents last revised July 25, 2018

To Build Truly Intelligent Machines, Teach Them Cause and Effect

Giovanni Vigna, Lastline

Tools and Methodologies for Pipework Inspection Data Analysis

Design and technology

AI 101: An Opinionated Computer Scientist s View. Ed Felten

SMART MANUFACTURING: A Competitive Necessity. SMART MANUFACTURING INDUSTRY REPORT Vol 1 No 1.

Towards a Consumer-Driven Energy System

Connecting Commerce. Professional services industry confidence in the digital environment. Written by

Counterspace Capabilities using Small Satellites: Bridging the Gap in Space Situational Awareness

Transcription:

WHITEPAPER ZoneFox Augmented Intelligence (A.I.) Empowering the Super-Human Element in Your Security Team

Introduction In 1997 Gary Kasperov, the chess Grandmaster, was beaten by a computer. Deep Blue, designed by IBM, had played Kasperov before in 1996 and beaten him in the first game of a six-game match eventually losing to Kasperov 4-2. After a year s worth of tweaking and heavy upgrades Deep Blue finally defeated Kasperov in a re-match. Although Deep Blue derived most of its power from brute-force analysis - it was, afterall, able to analyse 200 million potential moves per second. The story shocked the world because for the first time, human intelligence had been challenged - and surpassed - by a machine. The Brain s Last Stand, read the Newsweek headline. But in the years after his defeat to Deep Blue, Kasperov started experimenting with AI so he could study its impact on chess, and it was during this time that he formulated the Advanced Chess League, allowing humans to play with the assistance of machines. This later evolved into a freestyle chess event where it was found that the best results were obtained with a combination of both humans and computer assistance. This realisation that a combination of AI and human characteristics delivers the best insights is what drives the ongoing development of ZoneFox. The information security industry has been party to many a silver-bullet, but in recent years, with the growth of well-funded adversaries as well as the chronic and continuing data-breach issue, attitudes have changed in relation to being better prepared to identify and respond to an incident rather than simply defending against it. In this whitepaper we will provide you with an insight into how the latest developments in Machine Learning, combined with the ZoneFox Augmented Intelligence (A.I.) design philosophy can provide you with the insights you need to quickly and easily identify threats to your organisation - and empower your security team to act quickly and confidently. ZONEFOX AUGEMENTED INTELLIGENCE 2

A Changing Information Security Landscape When it comes to Information Security, historically we ve been good at stopping people from doing things, for example with access control, firewalls and so on - at least when they work, or have been configured properly! We ve also been OK at detecting commonly known attacks such as malware, certainly once they ve been around for a while. But we ve been really bad when it comes to providing insights into human behaviour or advanced malware. Why? Because of our traditional reliance on static rules. They re easy to subvert, and hard to update dynamically. SIEM came to the rescue to try and provide us with a greater insight into patterns of behaviour, but that too isn t as effective as we first thought. The class of solution that implements something called User Entity Behaviour Analytics (UEBA) has started to become particularly useful. User behaviour has been a hard problem due to the fact that human behaviour is erratic and hard to predict. People can be very sneaky about removing data from your organisation. Where UEBA becomes really useful is its ability to provide a level of flexible pattern recognition which rules are unsuitable for, and which humans simply can t achieve due to the excruciatingly large amount of data involved. And that s where ZoneFox steps up to the plate. ZONEFOX AUGEMENTED INTELLIGENCE 3

ZoneFox Augmented Intelligence Organizations use ZoneFox to assess and mitigate risk to their businesses, and in doing so, they follow the threat investigation pipeline: INFORMATION IDENTIFICATION EXPLORATION DISCOVERY ACTION TRIAGE PRESENTATION REPORT ZoneFox supports this pipeline through rule maintenance and alert monitoring, to case management and investigation, and finally reporting. At the moment, within an organization these activities are mostly manual, with rules having to be created and maintained, and the necessity to find related events not directly associated with an alert - something which can prove really difficult. Overall, this puts constraint on more detailed investigations and impacts breach detection and response time. Our solution is ZoneFox Augmented Intelligence. AI is the next generation in UEBA threat detection and investigation built using UBA on top of the existing ZoneFox platform. AI optimises each stage of the threat investigation pipeline and results in automated detection and blazingly fast investigation of anomalous behaviour. It s built on four pillars; Machine Learning, User Driven Data Mining Visualisation and Augmented Intelligence.. 1 Machine Learning People don t normally steal information, so AI performs UBA using machine learning to spot when a user s behaviour changes. It s really that simple. More specifically? AI analyses data across multiple dimensions and applies Bayesian probability metrics to produce complex mathematical models of user behaviour. These baselines allow AI to flag deviations from the norm in real time - allowing you to discover previously unknown anomalous behaviour. ZONEFOX AUGEMENTED INTELLIGENCE 4

To build these user profiles, AI observes historical and real time user data, until it has enough to accurately encompass normal behaviour. This is the training stage - which is fully automated, meaning no hand holding required. It also has a rapid turnaround time, allowing it to start detecting anomalous behaviour in just a few hours. However, turning people into mathematical models loses some information. With only this approach, AI could detect someone behaving abnormally - but not know why, or what exactly led up to it. We have to be able to take into account that a user could be behaving differently because, say, they ve switched roles in an organisation, or because they re taking data in advance of their resignation. But machine learning is a means to detect abnormal behaviour, not a path to complete autonomy. With just machine learning, you run the risk of drowning in a sea of un-actionable alerts. 2 User Driven Data Mining 2ccx Our brains are wired to pick out patterns and seek explanations for everything we encounter. By supporting human judgement and empowering our thought processes over artificial correlations, ZoneFox lets you prioritise what s important and relevant. AI scales with your organisation, allowing comprehensive investigation at every level of detail and encouraging exploration through simple, obvious interfaces and inherent discoverability. ZoneFox AI works on behalf of you, learning from the anomalies you find most valuable, and feeding this information back into its user behaviour models to screen out irrelevant detections - so as you learn from the system, the system learns from you. Coupled with machine learning, ZoneFox enables users to drill down and actively explore data complex relationships connecting multiple anomalous events - intuitively and fast. ZONEFOX AUGEMENTED INTELLIGENCE 5

3 Visualisation ZoneFox AI displays data so that you can rapidly prioritise high risk anomalies. Each visualisation clearly expresses the shape of the data, accentuating high risk anomalies while giving you a bird s eye view of user behaviour. And separate visualisations which share attributes are grouped together to allow fast correlation of disparate anomalies. We want our users to draw accurate conclusions faster through simple visual similarity and intuition. After all, this is where humans excel. Fig. 1 Risk intensity - weekly view 4 Augmented Intelligence Computers have always worked on behalf of users - and for the foreseeable future that s unlikely to change. They were made for algorithmic examination of data at scale. Humans have intuition: the ability to zero in on abstract conclusions. We can investigate while piecing together the reason behind a user s action. Teaching a computer how to decode someone s intentions is difficult, since our thought processes are non-algorithmic and differ wildly from person to person. No-one can understand human behaviour better than a human. AI is more than an artificial human substitute. We don t want to replace users, we want to empower them. We believe that the solution to detecting anomalous behaviour is a fusion of human and machine: Augmented intelligence. ZONEFOX AUGEMENTED INTELLIGENCE 6

About Zonefox ZoneFox is a market leader in User Behaviour Analytics that help secure your business-critical data - and backs it all up with a proven track record of protecting reputation, sales revenue, and competitive advantage. Bottom line? Really great Insider Threat protection that works the way your business needs it to, and not the other way round. Like to Learn More? www.zonefox.com/ai youtube.com/zonefoxvideo @zonefox alerts@zonefox.com Argyle House, Edinburgh, EH3 9DR 0845 388 4999