{"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":"markdown","source":"Generative AI\n\n   With the constant innovation in our heavily technology-dependent world, a recent yet highly impactful technology called generative AI will revolutionize nearly all aspects of life. Generative AI is ground-breaking because it can produce original content by analyzing a colossal amount of data instead of traditional machine learning algorithms, which are limited to a small range of outputs. Moreover, generative AI can save time when creating new content, boosting productivity significantly. Generative AI has been implemented in numerous fields, from medicine to education to full-stack development. Generative AI also can replace many jobs, as its ability to generate unique content can reduce the need for humans in such professions. Through generative AI’s capacity to create new human-like content, it can potentially spark significant change in occupations across the globe. \n    \n   The Large Language Model is a spectacular example of generative AI's potential for remarkable feats. Large Language Models are a type of machine-learning model that uses vast quantities of human-generated text to train the model to generate its natural language outputs. To achieve this goal, this model employs a process known as supervised fine-tuning. Initially, a pre-trained language model is trained using a transformer model as the underlying architecture of the model. The transformer model is used since it uses a self-attention mechanism by varying the importance assigned to different works. The model is then fine-tuned to boost its performance. For this base pre-trained model to become a large language model, its architecture must be scaled to work with a larger dataset. These large language models have numerous use cases, such as chatbots to assist with daily tasks to using natural language processing in customer service. Large Language Models embody the epitome of machine learning, using the power of AI to unlock the potential of machines to communicate with humans naturally. \n\n   An incredible example of generative AI is the image generator which uses deep learning technology to generate photo-realistic images. First, these image generators transform an input caption into Contrastive Language-Image Pretraining, which encodes the text into a vector representation. This vector representation’s dimensionality is then reduced, retaining the vital information. Following this, an image embedding is created using text embedding, which involves transforming the image into a vector. Finally, a diffusion model changes the image embedding into an actual image which is then scaled up. This image generation technology benefits many activities, such as saving time when creating advertising or creating video game graphics. The image generators we see today can already create stunningly realistic images and, with some improvement, can revolutionize industries by being able to fake images that would ordinarily require hundreds of hours of work.  \n\n   The implementation of generative AI into the workforce can create a massive loss of jobs which would need structures in place to ensure a smooth transition into this world of higher unemployment. When generative AI comes into the workforce, many jobs will be eliminated, but the jobs that will be the most impacted are jobs that require a medium amount of skill, not requiring an incredible amount of education, but not being a low-skill profession. In the lower-skill professions, though they can be automated, there is not much incentive to do so as there are a plethora of workers willing to do those jobs for a low salary. Many people not having a sustainable source of income will be a massive detriment to the economy and cause significant hardship to families without a job. To avoid overly severe impacts on the economy, some governmental structures can help reduce the effects of job loss on the economy. Though job replacement by AI will be hard on families and the economy, it enables people to work on things more fulfilling as their jobs are no longer needed for the economy to function. \n\n   Using generative AI technology to assist employees can not only improve employee efficiency but can also improve employee and customer satisfaction and is a much more effective way for employees to achieve their goals. This is exemplified by the effectiveness of having an AI assistant for technical support agents to help deal with customer interactions. When the technical support agents were paired with an AI assistant, they resolved problems significantly faster, fixed more problems, and the customer left the interaction feeling more satisfied (Table 1). Additionally, employees with AI assistance are far less likely to leave their positions, indicating higher employee satisfaction when they have an AI assistant. These benefits are most significant when the technology is implemented with higher-skill workers. Workers who adhere to the AI’s instructions show far better improvements indicating the AI’s competency when giving advice. These improvements are not limited only to technical support but can also significantly benefit employers in other fields. Though this study points to significant gains when implementing this technology, it is also crucial to consider the individual circumstance of the situation when executing it, as well as helping employees work with the AI assistant.\n\n   The different types of generative AI will revolutionize the modern world with their ability to generate original yet human-like content in seconds Across the world, students can benefit from the personalized learning experience offered by large language models, while the advertising industry can boost its productivity with image generators. Generative AI has already shown massive use in its ability to improve the efficiency of employees by granting them competent feedback in real time. Though generative AI has come a long way, it still has a long way to go before it can fully replicate what the human mind can do, but with consistent innovation, we should be there soon. \n\n\n\n\n\nWorks Cited\n\nAlderucci, Dean, et al. \"AI and the Rise of Patenting: Assessing the Economic Impact.\" Carnegie Mellon  University, Center for AI Analysis of Patents, 2023.\n\nBrynjolfsson, Erik, et al. “Generative AI at Work.” NBER Working Paper Series, no. 31161, \tNational Bureau of Economic Research, Apr. 2023.\n\nBrynjolfsson, Erik, et al. \"Quantifying the Distribution of Machine Learning’s Impact on Work.\"\t2023.\n\nGiacaglia, Giuliano. “Transformer.” Medium, 11 May 2023, towardsdatascience.com/transformer-141e32e69591.\n\nGuinness, Harry. “How Does Chatgpt Work?” Zapier, 2 June 2023, zapier.com/blog/how-does-chatgpt-work/. \n\nIslam, Arham. “How Do Dall·e 2, Stable Diffusion, and Midjourney Work?” MarkTechPost, 15 Nov. 2022,\nwww.marktechpost.com/2022/11/14/how-do-dall%C2%B7e-2-stable-diffusion-and-midjourney-work/. \n\nMcClelland, Calum. “The Impact of Artificial Intelligence - Widespread Job Losses.” IoT For All, 23 Feb. 2023, www.iotforall.com/impact-of-artificial-intelligence-job-losses. \n\n","metadata":{}},{"cell_type":"code","source":"import pandas as pd \n\ndf = pd.DataFrame([['','Using AI assistance', 'Not Using AI assistance'],['Average Handle Time',43,35],['Resolutions per Hour',2,2.5]])\n\nprint(df)","metadata":{"_kg_hide-input":true,"_kg_hide-output":true,"execution":{"iopub.status.busy":"2023-06-17T06:55:43.087311Z","iopub.execute_input":"2023-06-17T06:55:43.087962Z","iopub.status.idle":"2023-06-17T06:55:43.107753Z","shell.execute_reply.started":"2023-06-17T06:55:43.087905Z","shell.execute_reply":"2023-06-17T06:55:43.106137Z"},"trusted":true},"execution_count":null,"outputs":[]}]}