1/274
Looks like no tags are added yet.
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
Ransomware
A form of malware that is used to extort money from its victims. If the ransom is paid, cryptocurrency is usually used as it is difficult to trace.
Locker Ransomware
Cybercriminals locking a victim from their computer
Crypto Ransomware
Cybercriminals encrypting valuable files
Cryptoviral Extortion
Cybercriminals encrypting valuable files and the victim can get the decryption key if they pay the ransom.
WannaCry Ransomware
It was a cyber attack performed by the Shadow Brokers in 2017. This group of hackers took advantage of a weakness in the Microsoft Windows OS and using the hack called EternalBlue that they stole from the NSA, they exploited this weakness on computers that had not updated their OS with the patch for this weakness and used it to spread WannaCry. It affected more than 200,000 computers, affecting huge corporations and institutions like NHS and FedEx
Early AI
It was symbolic AI that used a knowledge-based system (or expert system) that contained a set of explicit facts and rules that were used to logically deduce or reason information. The disadvantage was its rigidness and inflexibility to produce different outcomes.
Botnets
A network of malware-infected computers (bots/zombies) controlled by the bot herder who communicates with them using the command-and-control (C&C) server that allows anonymity and remote programming.
3 Stages of Botnet (and Botnet types)
They start by hackers (1) finding and exploiting a vulnerability, then they (2) infect the device and then after infecting thousands or millions of bots they (3) activate it by controlling them to carry out further attacks i.e. DDOS, phishing and cryptojacking.
Why is it difficult to stop a Botnet?
It is difficult to stop a botnet because there are many bots and it is easy to create and expand the botnet.
Passive Infection
No human input i.e. exploited software vulnerability
Active infection
Due to human actions i.e. downloading an attachment or clicking a link
Malware
It is any malicious software including viruses and worms. It can retrieve sensitive information, steal computer resources and damage or disrupt computers and computer systems. It is composed of a payload which is the malicious code and the dropper which is the method used to transmit the payload.
Domain Generation Algorithm (DGA)
This generates many new domain names and cybercriminals use this to change the domains they use to launch the attacks which make it difficult to locate them.
Phishing
It uses social engineering to create unsuspecting emails and text messages that are fake but look legitimate that have a link or an attachment. This link can be used to download further malware, force the user to visit an infected site and make the user share their important login credential for cybercriminals to steal their money or data.
Spear phishing
Spear phishing is when the cybercriminal collects information on the individual or group (usually a group) and use that to make highly personalised phishing campaigns e.g. spam emails.
AI in (phishing)
This has made it easier for cybercriminals to make phishing campaigns more sophisticated and targeted e.g. correcting spelling mistakes and personalises messages.
Social Engineering
The term for doing malicious activity to manipulate users into allowing malware to infect their device that allows the cybercriminals to access many things i.e. sensitive information.
Ethical (white hat) hackers
Works under a code of ethics and for the benefit of the public
Grey Hat Hackers
Caught between the good and bad side and could be potentially hackers who have been rehabilitated
Black Hat Hackers
Operate on the wrong side of the law and usually make malware or phishing campaigns to gain financial benefits from their hacking activity
Cyberterrorists
These hackers have a specific target and definitely want to destroy targets and cause harm
Hacker
It is a person who uses computers to gain unauthorised access to data. There are different kinds of these people like the white hats, grey hats, black hats and cyberterrorist.
Ransomware-as-a-Service (RaaS)
Its a cybercrime business where ransomware developers sell ransomware to other hackers who are often amaterus and cannot decrypt the victim's files even if the victim is willing to pay.
Privacy
The ability of an individual or group (usually an individual) to hide themselves or information about themselves, revealing it selectively. What information do you want to reveal or hide?
Security
Having protection from or resilience against potential harm
Security vs. Privacy
There's an overlap between the two as to have privacy with your information, your information should be secure, protected and not stolen.
Social Media Attack: Fake Profiles
Cybercriminals impersonate legitimate users in order to collect information by fraudelent means. These are then used to manipulate stock prices, spread fake narratives and deceive users into giving out personal information or clicking on malicious links.
Social Media Attack: Compromised Accounts
When a cybercriminal gains access to a legitimate acount they can post malicious content, spread malware or impersonate the owner to defraud others. Large brands and their executives are usually targeted which make the impacts very severe.
Machine Learning (ML)
A branch of AI that mimics the way humans learn without explicit programming. It allows machines to learn from data and past experiences and the more it is exposed to data, its performance and accuracy in identifying patterns and making predictions and decisions are improved as it learns, develops and adapts. There are two types of machine learning: supervised and unsupervised machine learning.
Supervised Machine Learning
The machine is trained using a labelled dataset where the input and corresponding output are shown and so then after learning from many examples of them, the machine can make predictions (an output) when given a new input. It is used to predict outcomes.
Unsupervised Machine Learning
The machine is given an unlabelled dataset where it organises the data’s patterns, similarities and differences into groups. It is used to analyse and discover hidden patterns and similarities and differences.
Social Media Attack: Malicious Links and Content
Cybercriminals using social media to this and they are usually disguised as harmless posts or messages for unuspecting users. These can include phishing sites to steal login credentials or malware being downloaded onto the victim’s device.
Social Media Attack: Reconnaissance
Cybercriminals gathering information about a target through their social media before launching a cyber attack on them. This information can be easy-to-access and general information or personal details that can be used to craft targeted attacks i.e. spear phishing or identity theft.
Social Media Attack: Narrative Attacks
This attack is designed to manipulate the public perception of a usually big company in order to disrupt business operations and remove trust among customers to cause reputational and financial harm.
13 Australian Privacy Principles
Open and transparent management of personal info
Anonymity
Collection of solicited personal info
Dealing with unsolicited personal info
Notification of the collection of personal info
Use or discolosure of personal info
Direct marketing
Cross-border disclosure of personal info
Adoption, use or disclosure of government related identifiers
Quality of personal info
Security of personal info
Access of personal info
Correction of personal info
Cryptocurrency (+ Decentralised Transaction)
Digital currency that uses blockchain technology. Its transactions are decentralised, which means that it is controlled by all the memebers available on a peer-to-peer network rather than one authority like the bank. It’s transactions are faster, more secure, allow for anonymity and irreversible.
Blockchain Technology (+ Hashes)
Blocks that contain information stored chronologically in a chain as an freely open distributed digital ledger. Each block contains data, the block’s hash and the previous block’s hash. The hash is unique for each block and when a block is changed, that block’s hash changes so then all following blocks are invalid. The blockchain is distributed on a P2P network and everyone gets a copy of it. When someone creates a new block and participants on the network (usually miners) verify the transaction (proof-of-work) and it gets broadcasted to be added to the blockchain.
Miners (+ Proof-Of-Work)
They have two tasks (1) verify transactions through the proof-of-work (donce in Bitcoin) which is used to confirm and calculate block’s hashes which prevents tampering as it slows down new blocks being created and (2) mining/generating new Bitcoins. For their efforts, they are rewarded with Bitcoins.
What prevents blockchains from being tampered?
Hash, Proof-of-Work and Peer To Peer Network
GameOverZeus
This incident had 2 malwares: GameOverZeus and Crypto Locker. GameOverZeus is silent until you try to access your financial account and it sends your financial information to the criminals in Ukraine and Russia and they empty your account. Crypto locker locks all the files on your computer and they can only be unlocked when you pay the ransom before 72 hours, or you will lose that data. This botnet was distributed in a peer-to-peer fashion via an email with an attachment from your supposed friend. To protect yourself, update your Windows OS, buy Mcafee’s anti-virus and don’t open suspicious emails and attachments.
Mirai
This botnet in 2016 exploited IoT devices that still used the default factory username and passwords that were easy to target by scanning the internet for them. Once they were infected, they reported back to a C&C server. It used a DDOS attack that had 1.2TB of internet traffic that disrupted services i.e. Twitter, Reddit and Netflix. After this incident, manufacturers were encouraged to used unique default passwords and have automatic security updates while users were encouraged to change default passwords, keep firmware updated and have IoT devices on separate networks.
Mass Surveillance
The process of the federal and local governments and big tech companies using technologies and systems to unconsensually collect monitor and analyse information on individuals within a society. This is to maintain security, prevent threats and to have data for everyone so that they have data on people who may do wrongdoing in the future. They usually wiretap, track, directly mass intercept communications and more.
Why does user’s being aware of mass surveillance important?
This is important because it enables the potential for the abuse of power and control over individuals and violates the user’s privacy.
XKeyScore
A mass surveillance technology used by the NSA revealed by Edward Snowden. The NSA can search with no authorisation through a vast database of millions of individuals. It works by filling in an online form that is sent to the sensor networks around the world until it reaches the local database of metadata and content of the user (who can be anyone) they are trying to search for. The NSA could access emails, chats, browsing history, IP address and more. It was very successful as it helped capture over 300 terrorists.
What are fingerprints? (XKeyScore)
They enable NSA to track a user’s internet activity, regardless if they try to be anonymous i.e. where they have been on the web and who they’ve been talking to etc.
Tempora
A mass surveillance technology run by GCHQ. They placed data interceptors on transatlantic fibre-optic cables that carry out internet data in and out of UK, including internet traffic between the US and Europe. This allowed direct access to large amounts of global internet data, and they usually collected 21PB a day of data. GCHQ and NSA analysts looked through that content and meta content, which could be stored for up to 30 days.
Hacktivism
Using hacking techniques for political and/or social causes i.e. leveraging their skills to disrupt operations, spread information and/or provoke change. Criminals usually target governments using specific techniques i.e. website defacement, DDoS, data breaches, doxing and online shaming.
Website Defacement (Hacktivism)
Infiltrating a website to change its appearance, replacing it with their political message; like a digital graffiti
DDoS (Hacktivism)
Interupts operations, making them inaccessible and its a show of force
Data Breaches (Hacktivism)
Leaked information can cause brand damage
General Data Protection (GDPR)
A European Union Law announced in 2018 that is designed to harmonise data privacy laws. Its purpose is to regulate how organisations handle personal data and information. Non-compliance results in large fines and reputational damage. It is considered a progressive approach to data protection.
General Data Protection Regulation (GDPR) Scope
This regulatory law protects personal data which is any identifiable information i.e. the individual’s full name, location, IP address and cookies and also protects sensitive personal data i.e. race, political opinion, biometric data and health info.
Key Principle of GDPR: Purpose Limitation
The data processing purpose must be clear, open and aligned with individual’s expectations. So organisations must specify their purpose from the start, document purpose, inform individuals transparently and make any additional use is fair, lawful and clear.
Key Principle of GDPR: Data Minimisation
Organisations should only collect the data they actually need. They should identify the minimum necessary personal data to fulfill their purpose, which prevents over-collection. An example of this is an online store sign-up doesn’t need their political opinions.
Key Principle of GDPR: Storage Limitation
Personal data must not be kept for no longer than as needed especially if its in a form where you can identify the individual. Extended storage is allowed for public, research, or statistics but even then the individual’s rights must be protected. Also if user hasn’t used a website in a while, their data should be deleted
Key Principle of GDPR: Integrity and Confidentiality (Security)
Personal Data must be protected from unauthorised access, loss or damage so organisations must use security measures to prevent breaches. The regulatory law doesn’t define exact security protections but it requires proper access controls to the data e.g. encryption.
Lockbit 3.0
The most and recent advanced version of a ransomware that employs the RaaS model which means criminals can independently execute attacks globally. It encrypts sensitive files and demands ransom payments in exchange for the decryption key. The criminals can also threatento leak the stolen data and exploit the media to force them to pay the ransom. Further it excludes targets within Commonwealth to reduce penalty enforcements. Initially it can access the system via stolen log-in information (phishing or brute-force attack) or exploiting a software vulnerability and then the attacker can gain further access to the system laterally. To mitigate the ransomware, users are encourage to update their software, back up their data and use MFA.
Desktop Computing
It is the largest market for computers in terms of making money. It focuses on getting the best possible performance (optimisation) for the lowest price. It is important to customers and computer designers. It features high-performance and cost-reduced microprocessors.
Personal Mobile Devices (PMD)
It is a wireless device with a multimedia interface e.g. phone, tablets etc. The cost is a main concern so it is designed to be cost-effective. Application software are optimised for the web and media uses. It uses flash memory for energy and size efficiency while prioritising responsiveness, predictability and low power consumption.
Internet of Things (IoT) Devices / Embdedded Computers
It is embedded computers that are usually connected to the internet wirelessly. They use sensors and actuators that communicate and interact with the environment for smart applications i.e. smartwatches and thermostats. They come in the widest range of processing power and cost with price as a key factor in design.
Servers
They are the backbone of systems used by organisations. They are a computer or program that provides resources, data or services to other computers (clients) over a network. They act as a central hub, managing requests and enabling access to shared files, websites and applications. Important characteristics of them include: availability, scalability, efficiency and cost-effectiveness.
Cloud Service Level Agreement (SLA)
It is a legal contract between a cloud tenant (customer) and the service provider that specifies the provided services and the performance standard. It helps protect both parties to ensure transparent expectations and accountability by the service provider. It ensures that the provider meets the agreed service level and if services are not met the provider may face penalties (i.e. voided contract) or the contract renegotiated.
Ethical Issues in Cloud Computing: Data Confidentiality
The cloud service provider may access the user’s sensitive data so privacy is very important. However, your sensitive file is very unlikely to be compromised but law enforcement can still require the service provider to give the cloud tenant’s data to them.
Ethical Issues in Cloud Computing: Data Theft
The user’s data can be stolen which normally doesn’t happen with good service providers. However, if it does occur, the service provider may not report the incident to the cloud tenant.
Ethical Issues in Cloud Computing: Geographical Data Storage Issues
Law enforcement of the country where your cloud server is can demand the service provider to hand in your data.
Ethical Issues in Cloud Computing: Multi-tenancy Security Issues
Many cloud tenants can share storage and computational resources so a user may accidentally access someone else’s data. The service provider may not pay attention to the security of your data but they usually do encrypt your file when you upload it. This concept is used to maximised profits.
Ethical Issues in Cloud Computing: Transparency
The willingless of a cloud service provider to reveal the details of their security preparedness
Ethical Issues in Cloud Computing: Managerial Issues
Some cloud service providers may not have good, control, security and privacy management
Ethical Issues in AI: Transparency
AI algorithms are not often this because of copyright and trade secrets, making it hard to detect bias.
Ethical Issues in AI: Respect for Human Values
AI algorithms must uphold human values and promote individual well-being. This ensures sensitive to diverse cultures and beliefs while maintaining ethical integrity when having a positive societal impact.
Ethical Issues in AI: Bias
Since AI is always based on the dataset chosen by the research, it makes it impossible to create a truely neutral system.
Ethical Issues in AI: Privacy
Protecting the user’s privacy should be the highest priority so user permission must be gathered before collecting or using their data. This ensures the security and ethical handling of personal information.
Ethical Issues in AI: Accountability
AI decision-making must be trackable and reviewable (auditable) to ensure transparency especially when handling private and sensitive data i.e. biometrics and health records. This ensures responsibility and proprer monitoring in AI-driven processes.
Ethical Issues in Cloud Computing
Data Confidentiality and Theft, Geographical Data Storage, Multi-tenancy Security, Tranparency and Managerial issues
Ethics in Artificial Intelligence
Transparency, Bias, Respect for Human Values, Privacy and Accountability
Modern Computer (Von Neumann) Architecture and its 4 major subsystems
Memory, Input/output, the arithmetic/logic unit (ALU) and the control unit
Australia’s AI Ethics Framework
There are 8 AI Ethics Principles that ensure safe, secure and reliable AI. They aim to achieve safer, fairer and more reliable AI outcomes. They also reduce the risks and negative impacts of AI while helping businesses and organisations follow high ethical standards in AI design and use.
Benefits of Australia’s AI Ethics Principles
They include building public trust in AI products and organisations, increase consumer loyalty in AI-driven services and promotes positive outcomes from AI applications.
Tendencies Towards Privacy: Time
Users are usually worried about the current and/or future data than past data since its relevance has so-called expired.
Tendencies Towards Privacy: Requesting method
User is willing to manually give their information to their friends but not have these alerts to be sent automatically and frequently. Also the user should understand that, once their give their information away, it’s not just theirs.
Tendencies Towards Privacy: Extent
User may want to have their information reported ambigiously rather than very specific.
Integrity (GDPR)
Your data shouldn’t be altered
Confidentiality (GDPR)
Your information should be a secret
Cryptonomicon
It is a 1999 historical fiction novel by Neal Stephenson set during WW2 and the ‘90s.
Da Vinci Code
It is a 2003 novel by Dan Brown with a cryptographer female protagonist, Sophie Neveu.
Sneakers
It is a 1992 American thriller film directed by Phil Robinson. It is about Martin and his group of security specialists doing a job that proves to be nefarious with harsh consequences.
Snowden
It is a 2016 biographical film directed by Oliver Stone that is based on books. It deals with Snowden’s whistleblowing story about the mass surveillance conducted by the NSA
Mr. Robot
It is an American Drama Thriller TV series created by Sam Esmail. It stars Rami Malek as a cybersecurity engineer with many pyshcological disorders.
Deep Learning
A type of machine learning that attempts to copy the human brain. It uses an artificial deep neural network with artificial neurons that are software modules called nodes that use mathematical calculations that process data such as pictures, text and sounds to produce accurate output, insights and predictions. There are hundreds or thousands of neuron layers that work together to make decisions.
Generative AI
Works by identifying and encoding the patterns within huge existing data and then uses that information to understand the user’s language request to generate new and original content i.e. text, images, videos and audio etc as a response to a user prompt. There are different types of this e.g. Generative Adversarial Network (GANs) and Large Language Model (LLM).
Generative Adversarial Network (GANs)
Learns patterns in the input data to generate a new example based on the original data set. It uses two neural networks: a generator, which produces new content that resembles the original and realistic data, and the discriminator, which evaluates whether the data it receives is real or fake and this can improve through training.
Large Language Model (LLM)
A category of a foundation model that is capable of understanding and generating content in the human language. It requires a large amount of training data to process so that it can generate meaningful responses that are general and adaptable. It is used for tasks such as translation, question-answering and text completion tasks. An example of one is ChatGPT.
Operation Chanology
A hacktivist protest against the Church of Scientology. It was led by Anonymous and was launched in 2008, aiming to punish the Church for internet censorship as it censored an interview with Tom Crusie. Their methods include a DDoS attack and protests that attempted to expel the church from the internet and expose its practices.
Major Threats to Online Privacy: Hacking
Hackers break into computers to steal personal or financial data. FIles that are not encrypted are easy targets for hackers to have access to sensitve data. Stolen data can be exploited for further attacks e.g. phishing emails for new contacts. Online banking and business increase the risk of this, making cybersecurity important.
Major Threats to Online Privacy - Data Trading
User’s data such as their identity, interest and location can be traded as public property. This can be as a result of social media post being shared beyond the intended audiences. Third parties can buy and use this data, often without the user being aware as consent may be given through the terms and services agreements. So once you upload something on the internet, it is no longer fully yours as the data may remain online indefinitely.
Major Leak from WikiLeaks
In 2012, this organisation released classified communications between the U.S. State Department and foreign representatives. This included a video that showed a U.S. helicopter attack in Baghdad that killed journalists and civilians. This video raised ethical concerns about modern aerial warfare and military conduct. So the U.S. government condemened this service and forced payment services i.e. Amazon, PayPal, Visa and Mastercard to cut off donations to this service. Anonymous retaliated with DDoS attacks, temporarily shutting down those payment services.
Major Threats to Online Privacy: Tracking
Online tracking techniques i.e. cookies and pixel tracking can track and monitor browsing history, search history and the user’s time spent on sites etc. Social media and map apps also request location data. Most of this tracking is for ads, but sometimes tracking can be done by cybercriminals to carry out illegal activities. Mass surveillance is also large-scale tracking conducted by the government. Hence, tracking can compromise privacy, leading to potential misuse of this personal data or it can even end up in the wrong hands.
Cookies
They are small blocks of data that allow websites to track users visits and activities. They are embdedded in ads, web banners and videos and once a user consents to them, they are placed on their device and it monitors the user’s activity across websites which can last for less than a day to over a month. Third-party ones are generated by a separate domain from the one a user is currently visiting, which stores the information to present the user with targeted ads, depending on their activity.
Clusters and Warehouse-Scale Computers (WSCs)
It is a group of desktops or servers acting connected by local area networks (LANs) to act as a single system. The largest ones are called this and they consist of tens of thousands of severs that can act as one. They are designed for high availability, price-performance efficiency, and power optimisation. It is used in large-scale data centers for cloud computing and massive processing tasks.