{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"from IPython.display import Image\nimport os","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2023-07-16T22:18:57.080253Z","iopub.execute_input":"2023-07-16T22:18:57.080557Z","iopub.status.idle":"2023-07-16T22:18:57.111873Z","shell.execute_reply.started":"2023-07-16T22:18:57.080531Z","shell.execute_reply":"2023-07-16T22:18:57.110891Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Introduction\n\nArtificial intelligence is often thought of as a relatively new field however that is not the case. Artificial Intelligence has been a part of our lives for ages. However despite that, there is not one commonly consented definition of AI. This can be attributed to the fact that it is still an evolving field.\n\n\nWe are often unaware about the various ways in which AI has been helping us and making our lives easier for the past few years. A few examples of AI in our daily lives:\n\n1. Google maps: Shows the best route that we can take while reaching a destination and also makes sure to keep up with the live traffic to minimise time.\n\n2. Facial Recognition or Voice Recognition: Smart assistant devices like Amazon Alexa or Apple Siri that are triggered only with the voice or face registered are some of the most commonly known examples of AI. \n\n3. Automatic vehicles: Vehicles that do not require the interference of any human and can function just by scanning the route.\n\n4. Automatic household items: Like an automatic vacuum cleaner that cleans the entire house just by scanning.\n\nAs we have always heard that there are two sides to a coin. There are a lot of times where AI failed to do a task or gave wrong output. For example, several times google maps have come with wrong directions. The analogy of AI can be said to be similar to a household pet. The same way a pet learns from its owners and behaves in a similar way with others. So, if a pet is trained wrongly, it's going to have behavioural issues. Similarly, when AI is fed with wrong or faulty data it deviates from desired output and gives rise to ethical issues. For example, on march 23,2016 microsoft released an AI chatbot via twitter called 'Microsoft Tay. It caused controversy as the tweets that the bot was exposed to consisted of inflammatory languages, offensive languages or racial slurs and as a result it took that data and started tweeting something similar and hence had to be taken down in 16 hours. This led to the concern of where to limit the expansion of AI and how to control that ethically.","metadata":{}},{"cell_type":"markdown","source":"# Understanding Artificial Intelligence\n\nArtificial Intelligence (AI) is the ability of a computer or a robot controlled by a computer to do tasks that are usually done by humans because they require human intelligence and discernment. In other words, AI works on the rationale of reaction and behaviour of human beings and considers the patterns or reactions as input data and uses it to train itself. It tries to replicate the way a human being would react in a particular situation. For example, if there's a car that approaches us from the opposite side while driving, the reaction shown or the decision taken by a human being AI will try to do the same.\n\nAI has been growing exponentially in the past decades and it has already been touching our lives in unnoticeable ways. Everytime we do a google search an AI is being used to show the best possible result, everytime Siri is being asked a question, natural language processing and speech recognition is being used. Added with the potential  AI holds, it is probably going to be one of the biggest scientific breakthroughs humankind has ever made.","metadata":{}},{"cell_type":"markdown","source":"# Need for a set of rules\n*Used AI-generated content which was later edited by a human*\n\nAs our world becomes increasingly interconnected thanks to the rapid advancements in technology, with AI permeating every aspect of our lives as talked about above from our daily routines to complex algorithms for shopping experiences - the need for AI ethics has become paramount. With claims about us entering the AI spring and the continued evolution and integration of AI into society, it comes with its own set of challenges in terms of ethical development, deployment and consideration on its impact( as well as the unintended social consequences).  \nTo contextualise the importance of AI ethics, one needs to understand the immense transformative power that AI holds as a technology. AI has the potential to not just transform but even revolutionise industries ranging  from healthcare, to finance.  However great power comes with great responsibility. Without ethical guidelines in place, AI systems have been shown to  be prone to biases and if its deployment continues to be so fragmented, it’ll keep making the rich richer and poor poorer. \nAI ethics serve as the moral compass to ensure that AI technologies are developed and deployed in a manner that upholds fundamental principles with its underlying goal being to serve humankind’s good. \n","metadata":{}},{"cell_type":"markdown","source":"# AI ethics and development\n\nAI ethics is a set of guidelines that advise on the design and outcomes of Artificial Intelligence. Guiding principles on what should be allowed and trained to learn. It comes down to the ethical and moral values of the person designing the algorithm. For example, we can design a filter that will prevent the AI Models from learning violent and aggressive behaviour from human beings or ignore negative language and offensive speech. In real world examples if we see, google assistant, siri or alexa refrains the use of offensive language even if prompted by the user. This is where AI ethics came into action as the designers of the technology have used profanity filters while developing these technologies.\n\n**Ethical issues with AI**\n1. **Inequality**: In a capitalist society, inequality has always been a matter of great concern. There has been an ever increasing gap between the rich and the poor. Now, there are various jobs that AI has been taking up and overtime no manual intervention will be needed. For example, Twitter CEO has declared the introduction of a truck that will be automated and no driver will be required but trucking has been a large industry in USA or Canada. This means that a large section of the workforce will end up being unemployed. If there's no alternative employment provided to them it will lead to no income in that section of workforce and increase of income on the manufacturers side\n.\n2. **Bias & Discrimination**: Human beings come with all sources of cognitive biases, such as confirmation bias, and those inherent biases are exhibited in our behaviours and subsequently will be fed as data to the AI models. This will lead to the amplification of bias and discrimination on a larger scale.\n\n3. **Evil genies**: Imagine an AI system is asked to eradicate cancer in the world, and after a lot of computing, it comes up with a formula that eradicates cancer by killing people affected by it. Hence, this can be termed as unintended unforeseen circumstances.\n\n4. **Misuse** - Accelerated Hacking: AI increases the speed of what can be accomplished and in many cases, it exceeds the ability of human beings to follow along.\n\n5. **Singularity**: There's an argument by the researchers that by way of 'intelligence explosion', a self improving AI will become very powerful and humans will not be able to stop it from achieving its goals.\n","metadata":{}},{"cell_type":"markdown","source":"# Ethical Principles in AI development\n\n1. **Transparency**: The principle of transparency or the need to have a transparent process while developing AI technologies acts as a commitment that will increase interpretability, explainability or any other form of disclosure. Everyone is made aware of the way the technology has been designed to tackle issues in time of emergency or to be able to stop it when it malfunctions.\n\n2. **Justice and fairness**: The principle is expressed keeping in mind the justice and unwanted bias, and is a caution globally that AI might increase inequality and reinforce societal bias if not dealt carefully.\n\n3. **Responsibility and accountability**: This principle seems to be intertwined with justice and transparency, in a way that if both of these are promoted these will increase the sense of responsibility and accountability in the entities that will develop the AI technology. Responsibility basically means that any problems or flaws that are reflected by the AI technology or destruction done by the AI technology has to be compensated by the developers.\n\n4. **Privacy and data protection**: Privacy is not just a value but also a right, in such circumstances the data that is collected while developing AI models should be kept under high security and should not be misused.\n\n5. **Non-Maleficence**: Moral obligation as AI shouldn't be used as a way to harm others.","metadata":{}},{"cell_type":"markdown","source":"# Artificial Intelligence in healthcare\n\nArtificial intelligence can actually save lives and improve healthcare for millions of patients around the world. AI is helping personalise delivery of care, make hospitals more efficient and improve access to healthcare by providing accurate decision making tools. We are talking about having access to a computer model that is based on the experience of thousands of other patients, whether a treatment is likely to work and what works best on a particular patient based on their condition. No two people in the world are alike, but AI models help doctors learn about patients with similar conditions or similar genetic information, and make highly informed decisions about the diagnosis and their treatment options.\n\n**Use of AI in cancer paitients**\n\nCancer diagnosis can be immensely complicated, both for the doctors in making decisions about diagnosing a primary or secondary cancer, as well as for the patients in understanding the risks and success traits of the treatment option. There are AI models designed to streamline this process, by taking information from a number of sources. This involves feeding an AI model data from patients blood test, Xray images of suspected lesions as well as the genetic information from a tissue biopsy. The train AI model can rapidly consolidate this information and provide highly accurate prediction of the patient's diagnosis, treatment option most likely to succeed as well as the prognosis. There's an already available tool developed using AI and patient's genetic information that can accurately identify the primary site in case of cancer and hence empowers doctors to give patients a treatment that is known to work. These types of models can be expanded exponentially to predict accurate healthcare.\n\nThis means using an AI model to see if a certain population is more susceptible to a certain disease and whether they would respond more favourably to certain healthcare interventions. AI is giving us the ability to have a much more refined understanding of AI than we ever had\n\n**Ethics when using AI in healthcare**\n\nTraining an AI system requires both positive and negative outcomes fed back into the system. The subject of care must provide their informed consent and understand the experimental nature of the technology. The regulatory frameworks in place exist to help ensure that negative outcomes are reduced (based on experience) and that there is a responsible person in the decision making that led to the outcome. The responsibility should be taken care of by both the clinician and the engineer and these aspects should be established from a medico-legal, liability, education and socio-technical point of view of both","metadata":{}},{"cell_type":"markdown","source":"# Deepfakes and Generative AI for Disinformation and Deception\n\n**Generative AI:**\nGenerative AI is a subset of deep learning which means it uses artificial neural networks and processes both labelled or unlabelled data which uses supervised, unsupervised and semi-supervised methods. Generative AI can generate new data instances based on a learned probability distribution of existing data. Thus, generative models generate new content. The generative AI model which attempts to learn patterns on content for new content generation.\nThe generative AI process can take training code labelled data, unlabelled data of all data types and build a foundation model. For example, text, code, images, audio, video, etc.\nTwo types of Generative AI models exists: \n\n1. Gen AI language models that learn about patterns in language through training data. i.e. given some text it can predict what’s next.\n\n2. Gen AI image models that produce new images using techniques like diffusion, i.e. they can generate images from prompts.","metadata":{}},{"cell_type":"code","source":"Image(\"../input/images/Midjourney GenAi.jpg\")","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2023-07-16T22:18:57.116367Z","iopub.execute_input":"2023-07-16T22:18:57.117259Z","iopub.status.idle":"2023-07-16T22:18:57.165031Z","shell.execute_reply.started":"2023-07-16T22:18:57.117219Z","shell.execute_reply":"2023-07-16T22:18:57.164354Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"                Image 1: Images generated by 'midjourney' on text prompt","metadata":{}},{"cell_type":"markdown","source":"**Deepfakes:**\n Deepfake refers to a way in which artificial intelligence and machine learning techniques can be used to create manipulated digital media, mostly videos and the development of deepfake over the years makes it hard for the viewers to distinguish between what is real and fake. The term deepfake comes from “deep learning”( a subfield in AI/ML) and “fake”. Deepfakes alter existing images or videos by swapping faces, altering expressions, or manipulating speech. By training on a large data set of real images and videos the model learns how to depict the original source as close as possible.\nDeepfakes have existed for years now, but over the last year it has gained significant attention due to its misuse and the ethical concerns surrounding its existence. Recently deepfakes have been used to create realistic looking images or videos that falsely depict individuals doing things they never did. Although the creation of deepfake was with the intention of entertainment purpose, it is now maliciously used to create non-consensual explicit content of regular individuals and public figures. This has led to an increase in defamation and disinformation, and also raises questions on the trust on digital media files.\n\nEarlier in 2021, a tik tok account popped on everyone's explore page that posted videos of the famous hollywood actor Tom cruise goofing around in a men's clothing store, showing off a coin trick or even biting into a lollipop. As convincing as it looked to the viewers and audience but it wasn't actually Tom Cruise but actually a doppelganger of the same who was made to look and sound like him using artificial intelligence. The deepfakes were created by visual and AI effects artist Chris Umé with the help of a Cruise stand-in, actor Miles Fisher. This went on to be an example of the advancement and convincing power of deepfakes.\n","metadata":{}},{"cell_type":"code","source":"Image(\"../input/images/Tom cruise deepfake.png\")","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2023-07-16T22:18:57.166247Z","iopub.execute_input":"2023-07-16T22:18:57.166534Z","iopub.status.idle":"2023-07-16T22:18:57.209438Z","shell.execute_reply.started":"2023-07-16T22:18:57.166509Z","shell.execute_reply":"2023-07-16T22:18:57.208218Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"       Image 2: 'Tom Cruise' Deepfake on Tiktok by Miles Fisher and Chris Umé ","metadata":{}},{"cell_type":"markdown","source":"***Generative AI and disinformation:***\nInformation which is both inaccurate and spread with malicious intent is termed as disinformation. Disinformation can lead to panic in the general public and sometimes might even lead to the fueling of confirmation bias. Disinformation can be used in a campaign to tarnish images of brands and people alike. Generative AI can act as a great tool in creation of fake news and spread of disinformation among the common public. Like Winston Churchill said \"A lie gets halfway around the world before truth has a chance to put its pants on\". A classic example of generative AI being used for disinformation is the 'fake pentagon explosion' that took over social media with storm in May 2023. Within minutes, it was shared by numerous verified accounts without proper news verification and lead to amplification of the confusion.","metadata":{}},{"cell_type":"code","source":"Image(\"../input/images/pentagon explosion.png\")","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2023-07-16T22:18:57.210715Z","iopub.execute_input":"2023-07-16T22:18:57.211311Z","iopub.status.idle":"2023-07-16T22:18:57.234519Z","shell.execute_reply.started":"2023-07-16T22:18:57.211278Z","shell.execute_reply":"2023-07-16T22:18:57.233606Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"                    Image 3: Fake pentagon explosion generated by AI","metadata":{}},{"cell_type":"markdown","source":"A lot of Generative AI bots are available online that can create images based on text prompt and depending on the style of image chosen by the user. The image below is of a false explosion generated by a 'fotor' that is available online for free.","metadata":{}},{"cell_type":"code","source":"Image(\"../input/images/gen-ai india (2).png\")","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2023-07-16T22:18:57.236413Z","iopub.execute_input":"2023-07-16T22:18:57.236702Z","iopub.status.idle":"2023-07-16T22:18:57.310883Z","shell.execute_reply.started":"2023-07-16T22:18:57.236678Z","shell.execute_reply":"2023-07-16T22:18:57.308606Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"***Deepfakes for deception:***\nThis brings us to fake videos and deepfakes. Some of the early example of deepfakes ages like six years ago, things like inserting nicholas cage's face into different movies although it was not particularly dangeruous or convincing. The technology is also used to able to mimic people's voices and this has been used by cyber criminals in number of ases already and it will proliferate over the years. Deepfakes take disinformation to a new level by using the power of AI. The information campaigns are sophisticated which exploit vulnerability, spread information tailor-made to confuse the public and erode societal structures.\nRussia used deepfake of president zelensky during the time of Russia-Ukranian war asking the Ukranaian troops to surrender. Although, this deepfake was easier to spot and the spread of disinformation was prevented but this might not be the case always.","metadata":{}},{"cell_type":"markdown","source":"# Artificial Intelligence and Warfare\nSwarms of lethal drones with facial recognition that know your every move or unnamed armed robots that are near impossible to defeat, autonomous fighter jets that can travel at supersonic speeds and can withstand greater gravitational force than human pilot can survive, cyberattacks that incapacitate critical port infrastructure or disinformation campaigns and deepfakes during presidential elections, or even foreign adversaries taking out satellites, our eyes and ears in space rendering us blind to global events. All super intelligent weapons of terror. We are at the dawn of a new age of warfare.  The last era of warfare was the atomic bomb. I was keenly aware of how this technology has changed geopolitics and warfare. AI is one of the most critical technologies of our time and with deep implications of national security and democracy globally. As we saw in world war 2 in the atomic bomb, the country that is able to most rapidly and effectively integrate new technology in warfighting wins. It won’t be any different in the case of AI. \n\n**China vs USA:**\nFrom a technological perspective, China is already ahead of the USA in computer vision AI and in large language models like ChatGPT they are fast followers. In terms of military implications, China outspends everyone with an adjusted total military budget. China spends ten times more than the USA. \nThe first reason is Data supremacy. Despite having the largest fleet of military hardware in the world, most of the data of this fleet are thrown away or inaccessible, hidden away in HD that never see light of day. This is Achellies heel. In AI world, everything boils down to data, For defense AI, data from the internet isn’t enough. Most data has to come from military assests, sensor and collaboration with tech companies. Military commanders need to know how to use data as a military asset. \nSecond, despite being home to the leading tech companies at the forefront of AI, the USA tech industry has largely shied away from taking government contract.\n\n**Ukraine vs Russia:**\nThe Ukraine war has demonstrated that the nature of war has changed. Through AI overmatch Ukraine is challenging adversaries with far superior troops and weapons. Before the war, Russia spent an estimated 65 billion US dollars on military expenditure whereas UKraine spent an estimated six billion dollars only. It’s estimated that Russia had over 900,00 military troops and 1300 aircrafts whereas Ukraine only had 200,000 military troops and 130 aircrafts. Technologies such as drones, AI based targeting and image intelligence and javelin missiles enabled a shocking defence of Ukraine. AI is proving invaluable for defending ukrainian city and infrastructure against missiles and drone bombardment.\n\nIt’s clear that AI is increasingly powering warfare, and based on the rate of progress in the AI field, it can be said that in ten years it will be the dominant force.","metadata":{}},{"cell_type":"markdown","source":"# In depth into Generative Artificial Intelligence\n\n*While not AI generated content, we have taken help from ChatGPT to come up with the response and edited it according to how we deemed fit. It is however written by a human.*\n\nGenerative AI has been introduced as early as the 1960s however it was only in 2014 when they started getting popular with 2023 officially being crowned as the year for generative AI. However with its own set of benefits, comes its set of challenges. While we are still in the early phases, it is important to consider the ethical implications in its development and deployment.\n\n1. **Bias and Fairness:** Generative models are trained on vast amounts of datasets which may inadvertently contain implicit biases. It’s difficult to scan through them for the sheer amount of sizes and this may lead to generated content which reflects these biases and may contribute to their perpetuation. It’s important for ethical guidelines to be established to address the same.\n\n2. **Intellectual Property and Plagiarism:** one of the most highly debated topics in the last few months is\n    a. Firstly as mentioned before these models are trained upon vast amounts of data, some of which is obviously copyrighted data.  This raised a question on the content generated.\n    b. It also brings up who has the right to the AI generated content to deeper questions on if human creativity is something that is replaceable. Do machines also get IP rights ? \n    c. A balance needs to be struck which respects the intellectual property rights of the content owners and content. \n\n3. **Misinformation and Manipulation**: As discussed above, generative AI has the potential to create content which can be highly manipulated, the adverse effects of which has been discussed. However at the same time it’s important to be noted how like Open AI’s ChatGPT has often been reported for reporting facts confidently- which aren’t true. The fact that the system believes and reports it confidently is something which also needs to be addressed as this can lead to a great spread of misinformation especially because of the unawareness people hold about the same.\n\n4. **Privacy and data protection**: Data plays an integral role in the process of training these models and ensuring the protection of this sensitive data is paramount. At the same time, user consent and data collection needs to be governed. \n\n5. **Accountability and Transparency**: As of July 2023, with the recent charge against Open AI, it has made it clear that the development of such systems need to be a transparent process. This isn’t just because of the privacy and IP rights of individuals along with informing them as stakeholders of its possible biases but can also help in responsible deployment by building a transparent ecosystem where AI is accessible to all.","metadata":{}},{"cell_type":"markdown","source":"***EXAMPLES***\n\n1. **Manipulation and Misinformation:**\n    * One of the best examples of the most popular and terrifying deepfake is The Volodymyr Zelensky deepfake. While most are flagged off and are understood as jokes, this one has the Ukrainian President asking his soldiers to lay down his arms. At a sensitive time like a conflict, this holds extremely destructive potential to the lives of thousands if not millions of people. This is an example of the destructive potential deep fakes hold which will just get worse as the technology develops further [11].\n\n\n\n2. **Fairness and Bias Mitigation:**\n    * One of the better known cases, AI hiring tools have been displayed to be biassed against female candidates- being trained on historical data that favoured male candidates and hence perpetuating the bias further.\n    * ProPublica in 2016 analysed the rather well used COMPAS algorithm which was used to pass verdicts in US courts. It revealed the bias against black defendants and shedding light on the implicit biases in the system and how before these tools are officially adopted in the system, it is important to take stock of the situation to prevent social catastrophe.\n\n\n\n3. **Privacy and Data Protection:**\n    * The very fact that ChatGPT has been banned in Italy under it’s National Data Protection Authority is a testimony for the lack of data protection in the process [10]. \n    * A California law firm recently launched against Open AI, the creator of ChatGPT, and Microsoft a lawsuit for 3 billion dollars for violating the privacy and copyrights of millions of people by using data scraped from the internet to train their models [9].\n\n\n4. **Intellectual Property:**\n    * Sarah Silverman, a rather renowned US comedian and author of two books recently sued Open AI and Meta over a copyright infringement for training the models on her work without her explicit permission [7] as discussed previously, there are several such other cases such as Mona Awad and Paul Tremblay [8] for “unlawfully ingesting their book” being  recent examples.","metadata":{}},{"cell_type":"markdown","source":"# Suggestions\n\n1. **Transparency and Explainability**: Clear documentation and disclosure of data sources and decision making processes to create an informed environment with all stakeholders understanding its operations.\n\n2. **Fairness and Bias Mitigation**: Identify and establish benchmarks to fight against the biases and ensure equitable development and deployment of these AI systems. \n\n3. **Privacy and Data protection rights**: Establish a data protection environment which ensures ethical collection of data with informed consent of people.\n\n4. **Accountability in AI systems**: An ideally globally agreed upon standard mechanism needs to be set up to hold developers, organisations and AI systems responsible. ","metadata":{}},{"cell_type":"markdown","source":"# Conclusion\nWhile AI may not be the elixir which holds solutions to all the problems in the world, it is undeniably a key technology which can improve the life of humankind or make it worse. AI Ethics are hence instrumental to decide the foundation for the upcoming development for the technology which is paramount for our future. While this essay just provides an overview, it is important to acknowledge that a lot of work is being done by individuals, companies like Anthropic and  Governments with the EU recently launching its AI act. It’s an exciting time to be in and with the right amount of precaution, the future looks bright.","metadata":{}},{"cell_type":"markdown","source":"# References and Citations\n\n1. Vincent, J. (2016, March 24). Twitter taught Microsoft’s AI chatbot to be a racist asshole in less than a day. The Verge. https://www.theverge.com/2016/3/24/11297050/tay-microsoft-chatbot-racist\n\n2. Copeland, B. (2023, July 14). Artificial intelligence (AI) | Definition, Examples, Types, Applications, Companies, & Facts. Encyclopedia Britannica. https://www.britannica.com/technology/artificial-intelligence\n\n3. Ethics of artificial intelligence. (2023). UNESCO. https://www.unesco.org/en/artificial-intelligence/recommendation-ethics\n\n4. Office of the Director of National Intelligence. (n.d.). INTEL - Principles of Artificial Intelligence Ethics for the Intelligence Community. https://www.intelligence.gov/principles-of-artificial-intelligence-ethics-for-the-intelligence-community#:~:text=Respect%20the%20Law%20and%20Act,civil%20rights%2C%20and%20civil%20liberties.\n\n5. Cartage, C. (2020). Elon Musk moves to Texas and wants to change the trucking industry. Canal Cartage. https://canalcartage.com/news/elon-musk-moves-to-texas-and-wants-to-change-the-trucking-industry/\n\n6. Kachaamy, T., MD. (2023, May 4). Artificial intelligence and machine learning in cancer detection. Targeted Oncology. https://www.targetedonc.com/view/artificial-intelligence-and-machine-learning-in-cancer-detection\n\n7. Milmo, D. (2023b, July 11). Sarah Silverman sues OpenAI and Meta claiming AI training infringed copyright. Artificial Intelligence (AI) | the Guardian. https://amp.theguardian.com/technology/2023/jul/10/sarah-silverman-sues-openai-meta-copyright-infringement\n\n\n8. Creamer, E. (2023, July 5). Authors file a lawsuit against OpenAI for unlawfully ‘ingesting’ their books. Books | the Guardian. https://amp.theguardian.com/books/2023/jul/05/authors-file-a-lawsuit-against-openai-for-unlawfully-ingesting-their-books\n\n9. De Vynck, G. (2023, June 28). ChatGPT maker OpenAI faces a lawsuit over how it used people’s data. Washington Post. https://www.washingtonpost.com/technology/2023/06/28/openai-chatgpt-lawsuit-class-action/\n\n10. Bhati, D. (2023, April 5). Viral AI chatbot ChatGPT is banned in many countries, but why? Full list of countries. India Today. https://www.indiatoday.in/technology/news/story/viral-ai-chatbot-chatgpt-is-banned-in-many-countries-but-why-full-list-of-countries-2355938-2023-04-05\n\n11. Foley, J., & Guilcher, A. L. (2023). 20 of the best deepfake examples that terrified and amused the internet. Creative Bloq. https://www.creativebloq.com/features/deepfake-examples\n\n12. Lawton, G. (2023). What is generative AI? Everything you need to know. Enterprise AI. https://www.techtarget.com/searchenterpriseai/definition/generative-AI\n\n13. Wikipedia contributors. (2023). Midjourney. Wikipedia. https://en.wikipedia.org/wiki/Midjourney\n\n14. Wikipedia contributors. (2023a). Deepfake. Wikipedia. https://en.wikipedia.org/wiki/Deepfake\n\n15. Fisher, M. (2023, April 12). The Hollywood Reporter. The Hollywood Reporter. https://www.hollywoodreporter.com/feature/deepfake-tom-cruise-miles-fisher-1235182932/\n\n16. Edwards, B. (2023, May 23). Fake Pentagon “explosion” photo sows confusion on Twitter. Ars Technica. https://arstechnica.com/information-technology/2023/05/ai-generated-image-of-explosion-near-pentagon-goes-viral-sparks-brief-panic/\n\n17. Allyn, B. (2022, March 17). Deepfake video of Zelenskyy could be “tip of the iceberg” in info war, experts warn. NPR. https://www.npr.org/2022/03/16/1087062648/deepfake-video-zelenskyy-experts-war-manipulation-ukraine-russia\n\n18. Araya, D. (2022, July 18). AI is rapidly transforming warfare: new rules are urgently needed. Centre for International Governance Innovation. https://www.cigionline.org/articles/ai-is-rapidly-transforming-warfare-new-rules-are-urgently-needed/\n\n19. Liang, B. D. C. &. A. (2023, May 23). ChatGPT: Can China overtake the US in the AI marathon? BBC News. https://www.bbc.com/news/business-65034773\n\n20. Ahmed, Z. (2023, April 12). The AI arms race: How China and the US are competing in generative AI technology. The Indian Express. https://indianexpress.com/article/technology/artificial-intelligence/us-china-top-generative-ai-companies-8550693/\n\n21. Nanda, P. (2023, March 29). War In Ukraine By Far The Most ‘Hi-Tech’ The World Has Ever Seen; Experts Call It A ‘Field Day’ For American Firms. Latest Asian, Middle-East, EurAsian, Indian News. https://www.eurasiantimes.com/from-ai-internet-satellites-to-naval-drones-war-in-ukraine/\n\n22. Bloomberg, J. S. |. (2023, May 31). Ukraine war may become a proving ground for AI. Washington Post. https://www.washingtonpost.com/business/2023/05/31/ukraine-war-latest-us-can-bring-ai-quantum-computing-to-battle/63d221c4-ffa9-11ed-9eb0-6c94dcb16fcf_story.html\n","metadata":{}}]}