{"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":"#Connecting Files\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n  \n        ","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-07-08T23:19:15.410942Z","iopub.execute_input":"2023-07-08T23:19:15.411395Z","iopub.status.idle":"2023-07-08T23:19:15.450575Z","shell.execute_reply.started":"2023-07-08T23:19:15.41136Z","shell.execute_reply":"2023-07-08T23:19:15.44977Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n#importing dependencies\nimport torch\nimport matplotlib.pyplot as plt\nimport umap\nfrom sklearn.cluster import KMeans\nimport json\nfrom datetime import datetime, timedelta\n#connect data\narxiv_metadata_all = pd.read_json( \"/kaggle/input/2023-kaggle-ai-report/arxiv_metadata_20230510.json\",  lines=True, convert_dates=True)\n","metadata":{"execution":{"iopub.status.busy":"2023-07-08T23:19:15.45231Z","iopub.execute_input":"2023-07-08T23:19:15.452628Z","iopub.status.idle":"2023-07-08T23:22:23.307152Z","shell.execute_reply.started":"2023-07-08T23:19:15.452599Z","shell.execute_reply":"2023-07-08T23:22:23.305882Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Kaggle Essay on Generative AI\n\nI. Introduction \nArtificial Intelligence (AI) development worldwide, especially for developing countries has opened up opportunities with the rapid advancement of artificial intelligence.\nWhat is Generative AI?\nGenerative AI is a branch of artificial intelligence (AI) that aims to create new content by learning from existing data like images, text, videos, music, voice, or code. Generative AI can be used for various purposes including artistic expressions, data augmentation used for machine learning to split dataset for training, content creation, data synthesis which brings together results and examines the findings, and data anonymization which identifies, modifies, or removes personal information. Generative AI can enhance existing content like adding filters, modifying images to remove objects, altering faces, or changing backgrounds.\nArtificial intelligence leverage computers and machine learning to mimic the human mind in problem solving and decision-making while Data Science is the study of data to extract meaningful insights.\nThe purpose of this essay is to centralize and summarize the rapid advancement of AI in the past two years as it relates to Generative AI and the plethora of writeups within the Kaggle community and elsewhere.  It is my aim to prove that the advancement of Generative Ai has taken over the world with the potential of opening knowledge that will give anyone limitless opportunities if one taps into its data domain of knowledge and uses it to alleviate poverty.\n\nII. Automation by Generative AI \nIn the past two years, there have been several advancements in generative AI with advancements that have great potential in various areas, including creative arts, entertainment, and design. Generative AI can contribute to poverty alleviation efforts in the following ways:\nB. Examples of task automation: \nEmployment and Economic Opportunities: Generative AI can be used to create new job opportunities and stimulate economic growth. For instance, AI-generated content can fuel the creative industries by producing artwork, music, and literature, leading to increased employment and income generation for artists and creators.\n\nEducation and Skill Development: AI-powered educational tools and platforms can enhance learning experiences and provide access to quality education in remote and underserved areas. By leveraging generative AI, personalized content and interactive simulations can be created, improving the learning outcomes and skills of individuals, thereby increasing their employability and reducing poverty.\n\nData Analysis and Decision-Making: Generative AI techniques, such as generative adversarial networks (GANs), can be used for data synthesis and augmentation. This enables researchers, policymakers, and organizations to generate realistic synthetic data that can supplement real-world data, making it easier to analyze and make informed decisions. Accurate and timely decision-making is vital for poverty alleviation efforts.\n\nHealthcare and Medical Research: Generative AI can aid in medical research, drug discovery, and diagnosis. AI models can generate synthetic data to help researchers analyze and understand complex diseases, develop new treatments, and improve healthcare outcomes. Access to better healthcare and medical advancements can positively impact poverty reduction.\n\nResource Optimization: AI algorithms can help optimize resource allocation and improve efficiency in various sectors, such as agriculture, transportation, and energy. By using generative AI techniques, businesses and organizations can reduce waste, increase productivity, and create more sustainable practices. This, in turn, can contribute to economic growth and poverty reduction.\n\nC. Benefits and implications: \nGenerative AI can create new and original content or data based on large amounts of existing data, such as text, images, audio, video, or code. For example, generative AI can write poems (see appendix) , draw portraits, compose music, or generate code from natural language comments¹.\nGenerative AI is powered by foundation models, which are large neural networks that can learn from vast quantities of unstructured and unlabeled data in various formats. Foundation models can be used for a wide range of tasks and domains, such as natural language processing, computer vision, speech, and reinforcement learning².\nGenerative AI can be used to automate, augment, and accelerate work in various industries and fields, such as healthcare, finance, retail, media, and art. For example, generative AI can help diagnose diseases, design new drugs, optimize business processes, create marketing content, and generate new product ideas³.\n Generative AI can also be used for fun and entertainment purposes, such as creating art, music, games, and virtual worlds. For example, generative AI can produce realistic and diverse images from text descriptions (DALL-E), generate novel music from text or existing music snippets (Bard), or create entire virtual worlds from scratch (Midjourney)⁴.\nGenerative AI poses some challenges and risks for society and ethics, such as data quality, privacy, consent, ownership, accountability, transparency, and human dignity. For example, generative AI can produce fake news, deepfakes, misinformation, plagiarism, or harmful content that may affect people's trust, safety, rights, and values. Recently, I read an article where OpenAI is being sued because someone decided to research himself on AI and found that the data was totally inaccurate saying he did fraud. We will know the outcome of that case soon and if it will set some form of precedence going forward.\nData quality: Generative AI relies on large amounts of data to train its models. However, data may be scarce, biased, inaccurate, or outdated in some contexts. This may affect the performance and fairness of generative AI applications.\nEthical concerns: \nPrivacy: Generative AI may collect or use sensitive personal data without proper protection or permission.\nConsent: Generative AI may generate content or predictions without informing or obtaining consent from the users or subjects involved.\nOwnership: Generative AI may create content or predictions that infringe on intellectual property rights or cultural values.\nAccountability: Generative AI may cause harm or errors that are difficult to trace or remedy.\nTransparency: Generative AI may operate in opaque or complex ways that are hard to understand or explain.\n Human dignity: Generative AI may devalue or replace human creativity or agency.\nDigital divide: Generative AI may widen the gap between those who have access to digital technologies and those who do not. This may create or exacerbate inequalities in opportunities, resources, and outcomes.\n\nGenerative AI requires careful and responsible design, deployment, and governance to ensure that it is aligned with human values and social goals. Some of the possible actions that can help achieve this are:\nData governance: Establishing standards and protocols for data collection, storage, sharing, and use that ensure data quality, privacy, consent, and security.\nEthical principles: Developing and applying ethical principles and frameworks that guide the development and use of generative AI in accordance with human rights and social justice.\nStakeholder engagement: Involving and empowering diverse stakeholders, such as users, communities, experts, regulators, and civil society, in the co-creation and oversight of generative AI applications.\nEducation and awareness: Raising awareness and building capacity among different actors, such as developers, users, policymakers, and educators, on the benefits, risks, and best practices of generative AI.\nInnovation and inclusion: Fostering innovation and inclusion in the generative AI ecosystem by supporting research, development, entrepreneurship, and access for underserved groups and regions.\n\nAcademic articles on generative AI development in the past 2 years:\n\nGenerative AI funding exploded over the past 2 years¹: This article by Insider Intelligence analyzes the trends and challenges of generative AI funding, and how it can impact various industries and applications.\nWhat is ChatGPT, DALL-E, and generative AI?²: This article by McKinsey explains what generative AI is, how it works, and what kinds of problems it can solve. It also features some examples of generative AI tools like ChatGPT and DALL-E, and their potential implications for the future of work and creativity.\nGenerative AI Timeline: 9 Decades of Notable Milestones³: This article by CMSWire.com provides a historical overview of the development of generative AI, from its origins in the 1940s to its recent breakthroughs in the 2020s. It also highlights some of the key researchers and organizations that contributed to the field.\nThe brief history of artificial intelligence: The world has changed dramatically in the last decade⁴: This article by Our World in Data presents a visual summary of the progress and impact of artificial intelligence, including generative AI, in the last decade. It shows how AI systems have improved in tasks like image recognition, natural language processing, and image generation.\nGenerative AI: Privacy and tech perspectives⁵: This article by IAPP discusses some of the privacy and ethical issues that arise from the use of generative AI, especially in relation to personal data and identity. It also suggests some possible solutions and best practices for addressing these challenges.\n\n\nIII. Advanced Data Analysis with AI \nA: New paradigms of AI, such as self-supervised learning, meta-learning, and reinforcement learning, that enable models to learn from unlabeled data, adapt to new tasks, and optimize complex objectives.\nImage generation: AI image production has seen significant improvements in quality, diversity, and resolution. Some of the techniques used for image generation are generative adversarial networks (GANs), variational autoencoders (VAEs), style transfer, and latent diffusion.\nVideo generation: Text-to-video creation has also seen the emergence of generative videos that can produce realistic and diverse scenes from textual prompts. Some of the techniques used for video generation are video GANs, video VAEs, and neural rendering.\nAudio generation: AI can generate realistic and expressive speech, music, and sound effects from text or other inputs. Some of the techniques used for audio generation are WaveNet, WaveGAN, MelGAN, and Jukebox.\n3D generation: AI can generate realistic and detailed 3D models, scenes, and animations from text, images, or sketches. Some of the techniques used for 3D generation are 3D GANs, 3D VAEs, neural radiance fields (NeRFs), and differentiable rendering.\nHyper-personalization: AI can tailor content and experiences to individual preferences, needs, and contexts. Some of the applications of hyper-personalization are recommender systems, content creation, marketing, and education.\nFrom search to chat: AI can provide more interactive and conversational ways of accessing information and services. Some of the applications of chat are customer service, education, entertainment, and health care.\nMulti-modality: AI can integrate and generate content across different modalities, such as text, image, video, audio, and 3D. Some of the applications of multi-modality are content synthesis, summarization, translation, captioning, and question answering.\n\nB. Enhanced insights and predictions: \nGenerative AI-driven data analysis is a type of data analysis that uses algorithms and models that can generate new content based on existing data. It can help data scientists to make more accurate predictions and gain deeper insights into complex data by:\ni.\tAutomating repetitive tasks such as data collection, cleaning, tagging, and classification²⁴.\nii.\tGenerating software code to build analytic models².\niii.\tCreating realistic and diverse data sets for training and testing¹³.\niv.\tProducing novel and creative solutions for data problems¹³.\nv.\tEnhancing data exploration and visualization with natural language queries and image generation⁴.\nvi.\tContextualizing analytics results and providing persona-based insights².\n\nThe article Visualizing Global Attitudes Towards AI \n( Visualizing Global Attitudes Towards AI (visualcapitalist.com) shows a graph on how the world is thinking about AI and how developing countries see more hope in its advancement than developed countries.\n\n\nIV. My Observations\nI was doing some research recently, for a blog I was writing on ‘How to be an Effective Scrum Master in the Age of AI Development’ and found an AI company, Huggingface, who is transforming the world with the dissemination of its AI framework with its aim to democratize1 AI. I had the same thought when I started to research the potential of Open AI chatgpt and how quickly and to the point information was returned without the fluff. I call it my personal assistant. I could ask it to help me with whatever it is I thought of doing and it would spit out humanlike responses, that I could use almost entirely.  I understand that it, sometimes returns inaccurate information and requires follow up research. \nI am reminded of that movie I watched where this young man in an African country, made a windmill turbine electrical pump from trash materials and saved his family and neighbor and community from starvation. He had challenges getting access to information at first but eventually succeeded. Can you imagine the information this could provide some aspiring, passionate young individual with access to such information? It will provide a library for those who have limited access to information, education for those of limited means far and wide and quick information for the educated ones to improve.  I am quite ecstatic about this revolution. I see it reducing poverty around the world. Teach a man to fish, and you give him food for life, rather than give him fish for one meal and he becomes dependent. This access could bring nations around the world closer to the same level of development and less dependency. See movie clip here: https://youtu.be/nPkr9HmglG0.\nI found that researching with chatgpt AI tool brought back information that was clean and to the point. It was not crowded with distracting information that would draw my attention to other things. I find myself using the tool more often than I would use google, so I got the extension for chrome. This allowed me to glance at what information google returned and compare it to what chatgpt returned.  In my mind chatgpt was way easier to read or gather information quickly to take my next steps. I also, recently began compare results from Microsoft Bing AI tool and at first was not impressed, because it had a lot of distracting information but lately, I am very impressed with the new updates which also provided sources for its information and links without being distracting and providing helpful tips. Majority of this report was done with Bing Ai since chatgpt has data up to 2 years ago.\nOne thing that puzzled me while using the image processing AI tools, Midjourney and playground?  I observed that it did not return a dark-skinned woman if I asked it for a beautiful woman and we need to change that and train it to recognize a wide range of beauty. Also, it came back with multiple limbs, and extra body parts sometimes. So here is an area for improvements. The voice training AI tools are giving some much-improved results compared to a few years back. I was able to train my voice with Eleven Labs voice tool and get back wonderful results which I truly admire. With my trained voice I don’t need to do voice overs like a hundred times before I like the sound of it while editing.\nI saw a technology documentary where a company in Africa was using drones to do deliveries. They delivered medical supplies and returned with amazing accuracy and timely precision. See more here: https://www.youtube.com/watch?v=gVf9wMgrxms&t=41s\n\nV. Conclusion \nGenerative AI is a powerful and promising technology that can have a positive effect on human way of life especially, productivity but requires careful and responsible management to ensure that it is used for good and not evil. \nThe purpose of this essay to summarize the rapid advancement of AI in the past two years as it relates to Generative AI and the plethora of writeups within the Kaggle community and elsewhere. It proved that the advancement of Generative Ai has taken over the world with the potential of opening knowledge that will give anyone limitless opportunities if one taps into its data domain of knowledge  and backed by endless article, some AI written, was proven by the many articles attached in the index.\nAccording to the Forbs articles linked in the appendix, GeKaggle Essay - The Advancement in Generative AI between 2021 and 2023\n \nI. Introduction\nArtificial Intelligence (AI) development worldwide, especially for developing countries, has opened up opportunities with the rapid advancement of artificial intelligence.\nWhat is Generative AI?\nGenerative AI is a branch of artificial intelligence (AI) that aims to create new content by learning from existing data like images, text, videos, music, voice, or code. Generative AI is used for various purposes, including artistic expressions, data augmentation used for machine learning to split datasets for training, content creation, data synthesis, which brings together results and examines the findings, and data anonymization, which identifies, modifies or removes personal information. Generative AI can enhance existing content by adding filters, modifying images to remove objects, altering faces, or changing backgrounds.\nArtificial intelligence leverages computers and machine learning to mimic the human mind in problem-solving and decision-making, while Data Science is the study of data to extract meaningful insights.\nThis essay aims to centralize and summarize the rapid advancement of AI in the past two years as it relates to Generative AI and the plethora of writeups within the Kaggle community and elsewhere. I aim to prove that the advancement of Generative Ai has taken over the world with the potential of open knowledge that will give anyone limitless opportunities if one taps into its data domain and uses it to alleviate poverty.\n \nII. Automation by Generative AI\nIn the past two years, there have been several advancements in generative AI, with advances that have great potential in various areas, including creative arts, entertainment, and design. Generative AI can contribute to poverty alleviation efforts in the following ways:\nB. Examples of task automation:\nEmployment and Economic Opportunities: Generative AI can create new job opportunities and stimulate economic growth. For instance, AI-generated content can fuel the creative industries by producing artwork, music, and literature, leading to increased employment and income generation for artists and creators.\n \nEducation and Skill Development: AI-powered educational tools and platforms can enhance learning experiences and provide access to quality education in remote and underserved areas. By leveraging generative AI, personalized content and interactive simulations can be created, improving individuals' learning outcomes and skills, thereby increasing their employability and reducing poverty.\n \nData Analysis and Decision-Making: Generative AI techniques, such as generative adversarial networks (GANs), can be used for data synthesis and augmentation. Generative AI enables researchers, policymakers, and organizations to generate realistic synthetic data that can supplement real-world data, making it easier to analyze and make informed decisions. Accurate and timely decision-making is vital for poverty alleviation efforts.\n \nHealthcare and Medical Research: Generative AI can aid medical research, drug discovery, and diagnosis. AI models can generate synthetic data to help researchers analyze and understand complex diseases, develop new treatments, and improve healthcare outcomes. Access to better healthcare and medical advancements can positively impact poverty reduction.\n \nResource Optimization: AI algorithms can help optimize resource allocation and improve efficiency in various sectors, such as agriculture, transportation, and energy. With generative AI techniques, businesses and organizations can reduce waste, increase productivity, and create more sustainable practices.  \nC. Benefits and implications:\nGenerative AI can create new and original content or data based on large amounts of existing data, such as text, images, audio, video, or code. For example, generative AI can write poems (see appendix), draw portraits, compose music, or generate code from natural language comments¹.\nGenerative AI uses foundation models, large neural networks that can learn from vast quantities of unstructured and unlabeled data in various formats. Foundation models are tasks and domains, such as natural language processing, computer vision, speech, and reinforcement learning².\nGenerative AI can automate, augment, and accelerate work in various industries and fields, such as healthcare, finance, retail, media, and art. For example, generative AI can help diagnose diseases, design new drugs, optimize business processes, create marketing content, and generate new product ideas³.\n Generative AI is fun and entertainment purposes like creating art, music, games, and virtual worlds. For example, generative AI can produce realistic and diverse images from text descriptions (DALL-E), generate novel music from text or existing music snippets (Bard), or create entire virtual worlds from scratch (Midjourney)⁴.\nGenerative AI poses challenges and risks for society and ethics, such as data quality, privacy, consent, ownership, accountability, transparency, and human dignity. For example, generative AI can produce fake news, deepfakes, misinformation, plagiarism, or harmful content that may affect people's trust, safety, rights, and values. Recently, I read an article where Someone was suing OpenAI because someone decided to research himself on AI and found that the data was inaccurate, saying he committed fraud. We will know the case's outcome soon and if it will set some precedence going forward.\nData quality: Generative AI relies on large amounts of data to train its models. However, data may be scarce, biased, inaccurate, or outdated in some contexts. This may affect the performance and fairness of generative AI applications.\nEthical concerns:\nPrivacy: Generative AI may collect or use sensitive personal data without proper protection or permission.\nConsent: Generative AI may generate content or predictions without informing or obtaining permission from the users or subjects involved.\nOwnership: Generative AI may create content or predictions that infringe on intellectual property rights or cultural values.\nAccountability: Generative AI may cause harm or errors that are difficult to trace or remedy.\nTransparency: Generative AI may operate in opaque or complex ways that are hard to understand or explain.\n Human dignity: Generative AI may devalue or replace human creativity or agency.\nDigital divide: Generative AI may widen the gap between those with access to digital technologies and those without access. The data divide may create or exacerbate opportunities, resources, and outcomes inequalities.\n \nGenerative AI requires careful and responsible design, deployment, and governance to ensure alignment with human values and social goals. Some of the possible actions that can help achieve this are:\nData governance: Establishing standards and protocols for data collection, storage, sharing, and use that ensure data quality, privacy, consent, and security.\nEthical principles: Developing and applying ethical principles and frameworks that guide the development and use of generative AI following human rights and social justice.\nStakeholder engagement: For involving and empowering diverse stakeholders, such as users, communities, experts, regulators, and civil society, in the co-creation and oversight of generative AI applications.\nEducation and awareness: Raising awareness and building capacity among different actors, such as developers, users, policymakers, and educators, on the benefits, risks, and best practices of generative AI.\nInnovation and inclusion: Important to foster innovation and inclusion in the Generative AI ecosystem by supporting research, development, entrepreneurship, and access for underserved groups and regions.\n \nAcademic articles on generative AI development in the past two years:\n \nGenerative AI funding exploded over the past two years¹: This article by Insider Intelligence analyzes the trends and challenges of generative AI funding and how it can impact various industries and applications.\nWhat is ChatGPT, DALL-E, and generative AI?²: This article by McKinsey explains what generative AI is, how it works, and what kinds of problems it can solve. It also features some examples of generative AI tools like ChatGPT and DALL-E and their potential implications for the future of work and creativity.\nGenerative AI Timeline: 9 Decades of Notable Milestones³: This article by CMSWire.com provides a historical overview of the development of generative AI, from its origins in the 1940s to its recent breakthroughs in the 2020s. It also highlights some key researchers and organizations that contributed to the field.\nThe brief history of artificial intelligence: The world has changed dramatically in the last decade⁴. This article by Our World in Data summarizes visually the progress and impact of artificial intelligence, including generative AI, in the previous decade. It shows how AI systems have improved in image recognition, natural language processing, and image generation tasks.\nGenerative AI: Privacy and tech perspectives⁵: This article by IAPP discusses some of the privacy and ethical issues that arise from the use of generative AI, especially concerning personal data and identity. It also suggests some possible solutions and best practices for addressing these challenges.\n \n \nIII. Advanced Data Analysis with AI\nA: New paradigms of AI, such as self-supervised learning, meta-learning, and reinforcement learning, enable models to learn from unlabeled data, adapt to new tasks, and optimize complex objectives.\nImage generation: AI image production has significantly improved quality, diversity, and resolution. Some techniques used for image generation are generative adversarial networks (GANs), variational autoencoders (VAEs), style transfer, and latent diffusion.\nVideo generation: Text-to-video creation has also seen the emergence of generative videos that can produce realistic and diverse scenes from textual prompts. Some techniques used for video generation are video GANs, video VAEs, and neural rendering.\nAudio generation: AI can generate realistic and expressive speech, music, and sound effects from text or other inputs. Some of the techniques used for audio generation are WaveNet, WaveGAN, MelGAN, and Jukebox.\n3D generation: AI can generate realistic and detailed 3D models, scenes, and animations from text, images, or sketches. Some techniques used for 3D generation are 3D GANs, 3D VAEs, neural radiance fields (NeRFs), and differentiable rendering.\nHyper-personalization: AI can tailor content and experiences to individual preferences, needs, and contexts. Some hyper-personalization applications are recommender systems, content creation, marketing, and education.\nFrom search to chat: AI can provide more interactive and conversational ways of accessing information and services. Chat applications include customer service, education, entertainment, and health care.\nMulti-modality: AI can integrate and generate content across different modalities, such as text, image, video, audio, and 3D. Some multi-modality applications are content synthesis, summarization, translation, captioning, and question-answering.\n \nB. Enhanced insights and predictions:\nGenerative AI-driven data analysis is a type of data analysis that uses algorithms and models that can generate new content based on existing data. It can help data scientists to make more accurate predictions and gain deeper insights into complex data by:\n         i.            Automating repetitive tasks such as data collection, cleaning, tagging, and classification²⁴.\n       ii.            Generating software code to build analytic models².\n     iii.            Creating realistic and diverse data sets for training and testing¹³.\n     iv.            Producing novel and creative solutions for data problems¹³.\n    v.            Enhancing data exploration and visualization with natural language queries and image generation⁴.\n vi.            Contextualizing analytics results and providing persona-based insights².\n \nThe article Visualizing Global Attitudes Towards AI\n( Visualizing Global Attitudes Towards AI (visualcapitalist.com) shows a graph on how the world thinks about AI and how developing countries see more hope in its advancement than developed countries.\n  \nV. My Observations\nI was doing some research recently for a blog I was writing on ‘How to be an Effective Scrum Master in the Age of AI Development’ and found an AI company, Huggingface, that is transforming the world with the dissemination of its AI framework with its aim to democratize1 AI. I had the same thought when I started researching the potential of Open AI chatgpt and how quickly and to the point information was returned without the fluff. I call it my assistant. I could ask it to help me with whatever I thought of doing, and it would spit out humanlike responses that I could use almost entirely.  I understand that it sometimes returns inaccurate information and requires follow-up research.\nI remember watching a movie where this young man in an African country made a windmill turbine electrical pump from trash materials and saved his family, neighbor, and community from starvation. He had challenges getting access to information at first but eventually succeeded. Can you imagine the knowledge this could provide some aspiring, passionate young individuals with access to such information? It will provide a library for those with limited access to information, education for those of limited means, and comprehensive and quick tips for the educated ones to improve.  I am quite ecstatic about this revolution. I see it reducing poverty around the world. Teach a man to fish, and you give him food for life rather than fish for one meal, and he becomes dependent. This access could bring nations worldwide closer to the same level of development and less dependency. See the movie clip here: https://youtu.be/nPkr9HmglG0.\nResearching with the chatgpt AI tool brought back clean and to-the-point information and was not crowded with distracting details that would draw my attention to other things. I use the tool more often than Google, so I got the extension for Chrome. The browser extension allowed me to glance at what information Google returned and compare it to what chatgpt produced.  In my mind, chatgpt was easier to read or gather information quickly to take my next steps. I also recently began comparing results from the Microsoft Bing AI tool and was not impressed at first because it had a lot of distracting information. However, lately, I have been very impressed with the new updates, which also provided sources for its transmission and links without being distracting and helpful tips. I did this report with Bing AI since chatgpt had data up to 2 years ago.\nOne thing that puzzled me while using the image processing AI tools, Midjourney and Playground?  I observed that it did not return a dark-skinned woman if I asked it for a beautiful woman, and we need to change that and train it to recognize a wide range of beauty. Also, it came back with multiple limbs and extra body parts sometimes. So here is an area for improvement. The voice training AI tools are giving some much-improved results compared to a few years back. I was able to train my voice with the Eleven Labs voice tool and get back wonderful results which I truly admire. With my trained voiceover, I don’t need to do voiceovers a hundred times before I like the sound of it while editing.\nI saw a technology documentary where an African company used drones for deliveries. They delivered medical supplies and returned with amazing accuracy and timely precision. See more here: https://www.youtube.com/watch?v=gVf9wMgrxms&t=41s\n \nV. Conclusion\nGenerative AI is a powerful and promising technology that can positively affect the human way of life, especially productivity. Still, generative AI requires careful and responsible management to ensure it is for good and not evil usage.\nThe purpose of this essay is to summarize the rapid advancement of AI in the past two years as it relates to Generative AI and the plethora of writeups within the Kaggle community and elsewhere showed that the progress of Generative AI has taken over the world with the potential of opening knowledge that will give anyone limitless opportunities if one taps into its data domain of expertise and, backed by endless articles, some written by AI, proven by the many articles attached in the index.\nAccording to the Forbs articles linked in the appendix, Generative AI-driven data analysis is still an emerging technology. Still, it can transform how organizations use data for decision-making.\nHowever, it poses challenges and risks, such as data quality, accuracy, bias, ethics, and security. Therefore, human validation and oversight are essential to ensure the reliability and trustworthiness of generative AI outputs.\n \nChallenges with the Report.\nI enjoyed researching this article and summarizing the advancement of Generative AI for others to have fun reading. I would love to display numbers and charts from the analysis and dataset. I had challenges with the dataset being too large locally, CPU freezing on Kaggle issues, and connection issues with Colab recognizing the.json file. I will continue to work on this, but the next step would be to run more stats.\nFinally, Generative AI can alleviate poverty with the help of AI advancement. Which AI will be the leader soon, Bing AI, Chatgpt, or Google?\n \n\n \nAppendix\nAdditional Reading\n¹: Top 10 Generative AI Trends That Will Be Prevalent in 2023\n²: What is ChatGPT, DALL-E, and generative AI? | McKinsey\n³: Generative AI Is Exploding. These Are The Most Important Trends You Need To Know - Forbes\n⁴: Generative AI: Privacy and tech perspectives\n⁵: The History of Artificial Intelligence from the 1950s to Today\n⁶: A Brief History of Generative AI - Medium\n⁷: BUSINESS FIGHTS POVERTY GENERATIVE AI AND SOCIAL IMPACT: THE ROLE OF BUSINESS\n⁸: How AI Can Help Fight Poverty | Dell USA\n⁹: Artificial Intelligence and Poverty - The Borgen Project\n[^10^]: Artificial Intelligence (AI) and Poverty Reduction in the Fourth Industrial Revolution (4IR) Era: A Systematic Review\n¹¹: AI Is a Game-Changer in the Fight Against Hunger and Poverty. Here's Why - NBC News\nScholar Articles\n¹: What is ChatGPT, DALL-E, and generative AI? | McKinsey\n²: Innovation Insight for Artificial Intelligence Foundation Models\n³: Top 70+ Generative AI Applications / Use Cases in 2023 - AIMultiple\n⁴: Infographic: Generative AI Explained by AI - Visual Capitalist\n: Generative Artificial Intelligence: Trends and Prospects | IEEE Journals & Magazine | IEEE Xplore\n \nSource: Conversation with Bing\n(1) What every CEO should know about generative AI | McKinsey. https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/what-every-ceo-should-know-about-generative-ai.\n(2) What is ChatGPT, DALL-E, and generative AI? | McKinsey. https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai.\n(3) Generative AI defined: How it works, benefits and dangers - TechRepublic. https://www.techrepublic.com/article/what-is-generative-ai/.\n(4) Infographic: Generative AI Explained by AI - Visual Capitalist. https://www.visualcapitalist.com/generative-ai-explained-by-ai/.\n \nSource: Conversation with Bing, 7/4/2023\n(1) Top 10 Generative AI Trends That Will Be Prevalent in 2023. https://www.analyticsinsight.net/top-10-generative-ai-trends-that-will-be-prevalent-in-2023/.\n(2) What is ChatGPT, DALL-E, and generative AI? | McKinsey. https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai.\n(3) Generative AI Is Exploding. These Are The Most Important  - Forbes. https://www.forbes.com/sites/konstantinebuhler/2023/04/11/ai-50-2023-generative-ai-trends/.\n(4) Generative AI: Privacy and tech perspectives. https://iapp.org/news/a/generative-ai-privacy-and-tech-perspectives/.\n(5) The History of Artificial Intelligence from the 1950s to Today. https://www.freecodecamp.org/news/the-history-of-ai/.\n(6) A Brief History of Generative AI - Medium. https://matthewdwhite.medium.com/a-brief-history-of-generative-ai-cb1837e67106.\n(7) BUSINESS FIGHTS POVERTY GENERATIVE AI AND SOCIAL IMPACT: THE ROLE OF .... https://businessfightspoverty.org/wp-content/uploads/2023/06/Generative-AI-and-Social-Impact-Role-of-Business.pdf.\n(8) How AI Can Help Fight Poverty | Dell USA. https://www.dell.com/en-us/perspectives/how-ai-can-help-fight-poverty/.\n(9) Artificial Intelligence and Poverty - The Borgen Project. https://borgenproject.org/tag/artificial-intelligence-and-poverty/.\n(10) Artificial Intelligence (AI) and Poverty Reduction in the Fourth .... https://www.preprints.org/manuscript/202009.0362/v1.\n(11) AI Is a Game-Changer in the Fight Against Hunger and Poverty ... - NBC News. https://www.nbcnews.com/mach/tech/ai-game-changer-fight-against-hunger-poverty-here-s-why-ncna774696.\n \nBing AI - Recent scholarly article resources\n(1) Generative AI funding exploded over the past two years. https://www.insiderintelligence.com/content/generative-ai-funding-exploded-over-past-2-years.\n(2) What is ChatGPT, DALL-E, and generative AI? | McKinsey. https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai.\n(3) Generative AI Timeline: 9 Decades of Notable Milestones - CMSWire.com. https://www.cmswire.com/digital-experience/generative-ai-timeline-9-decades-of-notable-milestones/.\n(4) The brief history of artificial intelligence: The world has changed .... https://ourworldindata.org/brief-history-of-ai.\n(5) Generative AI: Privacy and tech perspectives. https://iapp.org/news/a/generative-ai-privacy-and-tech-perspectives/.\nSource: Conversation with Bing\n(1) How Could Generative AI Impact The Data Analytics Landscape? - Forbes. https://www.forbes.com/sites/forbesbusinesscouncil/2023/05/24/how-could-generative-ai-impact-the-data-analytics-landscape/.\n(2) Generative AI for Powerful and Seamless Data Analysis | Dremio. https://www.dremio.com/generative-ai/.\n(3) Generative AI and the Future of Decision Intelligence. https://www.aberdeen.com/big-datapro-essentials/generative-ai-and-the-future-of-decision-intelligence/.\n(4) Generative AI: What Is It, Tools, Models, Applications and Use Cases. https://www.gartner.com/en/topics/generative-ai.\nData Analysis for DataScience\n \nnerative AI-driven data analysis is still an emerging technology, but it has the potential to transform the way organizations use data for decision-making. \nHowever, it also poses some challenges and risks, such as data quality, accuracy, bias, ethics, and security. Therefore, human validation and oversight are essential to ensure the reliability and trustworthiness of generative AI outputs.\n\nChallenges with the Report.\nI enjoyed researching this article and summarizing the advancement of Generative AI for others to have fun reading and hope it was informative. I would love to have some more time to display numbers and charts from the analysis and dataset given. I had challenges with the dataset being too large locally, CPU freezing on Kaggle issues and connection issues with Colab recognizing the .json file which I have not resolved yet. I will continue to work on this, but the next step would be to run more stats. \nFinally, I think poverty can be alleviated with the help of AI advancement. Which do you think will be the leader soon, Bing AI, Chatgpt or Google?\n\n \nSources - Citations:\n\nAdditional Reading\n¹: Top 10 Generative AI Trends That Will Be Prevalent in 2023\n²: What is ChatGPT, DALL-E, and generative AI? | McKinsey\n³: Generative AI Is Exploding. These Are The Most Important Trends You Need To Know - Forbes\n⁴: Generative AI: Privacy and tech perspectives\n⁵: The History of Artificial Intelligence from the 1950s to Today\n⁶: A Brief History of Generative AI - Medium\n⁷: BUSINESS FIGHTS POVERTY GENERATIVE AI AND SOCIAL IMPACT: THE ROLE OF BUSINESS\n⁸: How AI Can Help Fight Poverty | Dell USA\n⁹: Artificial Intelligence and Poverty - The Borgen Project\n[^10^]: Artificial Intelligence (AI) and Poverty Reduction in the Fourth Industrial Revolution (4IR) Era: A Systematic Review\n¹¹: AI Is a Game-Changer in the Fight Against Hunger and Poverty. Here's Why - NBC News\n\nAcademic/Scholarly Articles\n¹: What is ChatGPT, DALL-E, and generative AI? | McKinsey\n²: Innovation Insight for Artificial Intelligence Foundation Models\n³: Top 70+ Generative AI Applications / Use Cases in 2023 - AIMultiple\n⁴: Infographic: Generative AI Explained by AI - Visual Capitalist\n: Generative Artificial Intelligence: Trends and Prospects | IEEE Journals & Magazine | IEEE Xplore\n\nSource: Conversation with Bing\n(1) What every CEO should know about generative AI | McKinsey. https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/what-every-ceo-should-know-about-generative-ai.\n(2) What is ChatGPT, DALL-E, and generative AI? | McKinsey. https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai.\n(3) Generative AI defined: How it works, benefits and dangers - TechRepublic. https://www.techrepublic.com/article/what-is-generative-ai/.\n(4) Infographic: Generative AI Explained by AI - Visual Capitalist. https://www.visualcapitalist.com/generative-ai-explained-by-ai/.\n\nSource: Conversation with Bing, 7/4/2023\n(1) Top 10 Generative AI Trends That Will Be Prevalent in 2023. https://www.analyticsinsight.net/top-10-generative-ai-trends-that-will-be-prevalent-in-2023/.\n(2) What is ChatGPT, DALL-E, and generative AI? | McKinsey. https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai.\n(3) Generative AI Is Exploding. These Are The Most Important ... - Forbes. https://www.forbes.com/sites/konstantinebuhler/2023/04/11/ai-50-2023-generative-ai-trends/.\n(4) Generative AI: Privacy and tech perspectives. https://iapp.org/news/a/generative-ai-privacy-and-tech-perspectives/.\n(5) The History of Artificial Intelligence from the 1950s to Today. https://www.freecodecamp.org/news/the-history-of-ai/.\n(6) A Brief History of Generative AI - Medium. https://matthewdwhite.medium.com/a-brief-history-of-generative-ai-cb1837e67106.\n(7) BUSINESS FIGHTS POVERTY GENERATIVE AI AND SOCIAL IMPACT: THE ROLE OF .... https://businessfightspoverty.org/wp-content/uploads/2023/06/Generative-AI-and-Social-Impact-Role-of-Business.pdf.\n(8) How AI Can Help Fight Poverty | Dell USA. https://www.dell.com/en-us/perspectives/how-ai-can-help-fight-poverty/.\n(9) Artificial Intelligence and Poverty - The Borgen Project. https://borgenproject.org/tag/artificial-intelligence-and-poverty/.\n(10) Artificial Intelligence (AI) and Poverty Reduction in the Fourth .... https://www.preprints.org/manuscript/202009.0362/v1.\n(11) AI Is a Game-Changer in the Fight Against Hunger and Poverty ... - NBC News. https://www.nbcnews.com/mach/tech/ai-game-changer-fight-against-hunger-poverty-here-s-why-ncna774696.\n\nBing AI - Recent scholar article resources\n(1) Generative AI funding exploded over the past 2 years. https://www.insiderintelligence.com/content/generative-ai-funding-exploded-over-past-2-years.\n(2) What is ChatGPT, DALL-E, and generative AI? | McKinsey. https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai.\n(3) Generative AI Timeline: 9 Decades of Notable Milestones - CMSWire.com. https://www.cmswire.com/digital-experience/generative-ai-timeline-9-decades-of-notable-milestones/.\n(4) The brief history of artificial intelligence: The world has changed .... https://ourworldindata.org/brief-history-of-ai.\n(5) Generative AI: Privacy and tech perspectives. https://iapp.org/news/a/generative-ai-privacy-and-tech-perspectives/.\nSource: Conversation with Bing\n(1) How Could Generative AI Impact The Data Analytics Landscape? - Forbes. https://www.forbes.com/sites/forbesbusinesscouncil/2023/05/24/how-could-generative-ai-impact-the-data-analytics-landscape/.\n(2) Generative AI for Powerful and Seamless Data Analysis | Dremio. https://www.dremio.com/generative-ai/.\n(3) Generative AI and the Future of Decision Intelligence. https://www.aberdeen.com/big-datapro-essentials/generative-ai-and-the-future-of-decision-intelligence/.\n(4) Generative AI: What Is It, Tools, Models, Applications and Use Cases. https://www.gartner.com/en/topics/generative-ai.\nData Analysis for DataScience\n\n \nThis is a Poem for July 4th I Generated Using Chatgpt\n \nIn the land of stars and stripes unfurled,\nWhere freedom's spirit lights the world,\nOn this day of celebration, we find,\nA nation united, heart and mind.\n\nOn July Fourth, we stand as one,\nRecalling battles fought and won,\nA declaration, bold and brave,\nInscribed in history, our liberty engraved.\n\nIndependence Day, a hallowed name,\nA symphony of fireworks, love aflame,\nWe honor those who paved the way,\nWith courage, vision, and hope's array.\n\nFrom sea to shining sea, we declare,\nOur sovereign souls, we proudly bear,\nWith liberty's torch, we light the path,\nGuided by justice, love's aftermath.\n\nIn every state, in every town,\nThe American dream, forever renowned,\nWe hold these truths, both old and new,\nIn our hearts, freedom's fire will forever ensue.\n\nWith voices lifted, our anthem we sing,\nA melody of unity, hope's wings,\nIn this great land, the home we share,\nWe cherish the freedom that's beyond compare.\n\nSo on this day, with joyous pride,\nIn celebrations far and wide,\nWe stand as one, united and strong,\nIn the pursuit of rights we long.\n\nOn Independence Day, we salute,\nThe stars and stripes, resolute,\nFor in our hearts, the fire will burn,\nFor freedom's cause, we'll always yearn\n","metadata":{}},{"cell_type":"code","source":"arxiv_metadata_all.head()","metadata":{"execution":{"iopub.status.busy":"2023-07-08T23:22:23.308746Z","iopub.execute_input":"2023-07-08T23:22:23.309101Z","iopub.status.idle":"2023-07-08T23:22:23.354862Z","shell.execute_reply.started":"2023-07-08T23:22:23.30907Z","shell.execute_reply":"2023-07-08T23:22:23.353484Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"arxiv_metadata_all.describe()","metadata":{"execution":{"iopub.status.busy":"2023-07-08T23:22:23.357389Z","iopub.execute_input":"2023-07-08T23:22:23.357758Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"len(arxiv_metadata_all)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Convert the date column to datetime if it's not already in that format\narxiv_metadata_all['date'] = pd.to_datetime(arxiv_metadata_all['date'])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Filtering for the past 2 years","metadata":{}},{"cell_type":"code","source":"# Calculate the date two years ago from today\ntwo_years_ago = datetime.now() - timedelta(days=2*365)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Filter rows based on the conditions\nfiltered_rows = arxiv_metadata_all[(arxiv_metadata_all['date'] > two_years_ago) & (arxiv_metadata_all['categories'].str.contains('generative AI', case=False))]","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Print the filtered rows\nprint(filtered_rows.head())","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}