https://everyholiday.netlify.app/
This repositories is created for Writing45 imaginary project website development.
This is a static website deployed on the internet, can be accessied by anyone via a public URL. The code is hosted on GitHub and deployed automatically by Netlify, every code will update refreshes the site.
- Publicly accessible — Anyone on any device can browse it by opening the link, with no file download.
- Automated deployment — After pushing code to GitHub, Netlify automatically builds and publishes the latest version — no manual uploads.
- The GitHub repository stores all source code (HTML, PNG images, etc.) and records every change.
- The repo is linked to Netlify: whenever new code is pushed to GitHub, Netlify pulls and deploys it automatically.
- A visitor opens the Netlify URL, and the server returns the page over HTTPS for the browser to render.
- Static pages: HTML / CSS / image assets (PNG)
- Code hosting: GitHub
- Deployment: Netlify(continuous deployment)
When describing my greatest concern about the future, I am reminded of the many catastrophic events throughout history that were caused by the spread of false information. On the battlefield of the Battle of Fei River in 383 CE, the commander of the Jin asked the Qin forces to retreat one hundred kilometers so the two sides could formally engage after crossing the river. However, during the retreat, soldiers in the rear of the powerful Qin army mistakenly believed that the front line had already been defeated, and this misinformation caused panic and mass desertion. From that point, the Qin dynasty declined. Similar incidents have appeared repeatedly throughout world history before and after the Battle of Fei River: the panic surrounding Nostradamus, the Y2K scare, and the widespread rumors of local monsters that circulated during the last century. Errors and distortions in information can create social panic, and in the modern era, with the development of technology and the internet, even minor misinformation can spread at a much faster rate. Everyone has equal access to information, but not everyone possesses an equal ability to distinguish truth from falsehood. Those who trust false information stand firmly together, raised ostraca casting “truth” into exile.
What will further accelerate this problem is the generative AI controlled by capitalists. The development and iteration of generative AI in recent years have been unstoppable. It can efficiently collect data, organize text, and provide information to users, leading to an increasing number of people integrating generative AI into their daily lives. However, AI also poses a series of risks that remain uncertain and continuously evolving. Firstly, AI algorithms and models are often regarded as “black boxes.” (Zewe) The complexity of AI systems makes it extremely difficult for people to understand why these systems arrive at specific conclusions or how they generate predictions. When AI produces incorrect information, this opacity can conceal the dangers embedded within the system, making it difficult for people to respond critically. Secondly, Kush Varshney, a distinguished research scientist and senior manager at IBM Research, stated in the video “AI Trust, Transparency and Governance” released by the IBM AI Academy: “If we cannot trust these models, we cannot truly benefit from AI technology.”(Kush) In other words, users must place a certain degree of trust in AI systems in order to use them effectively. Yet when people become overly dependent on generative AI, the opacity of these systems and the unreliability of their information sources create even greater problems. AI-generated misinformation, such as incorrect scientific illustrations of animals, fabricated historical records, or distorted references to books and sources, can easily circulate online. This seemingly credible but distorted content contains not only factual inaccuracies but also fabricated information that may cause real harm.
Therefore, my imitation project was initially designed around a future world in which AI monopolized the source of information and, as the sole trusted authority, caused the collective collapse of society after making a single accidental mistake. I found this idea compelling because in current interactions between generative AI and the public, most errors produced by AI still appear unconscious or unintentional. When these systems encounter people who lack background knowledge and are unwilling to seek authoritative sources, misinformation can spread extremely easily on a large scale. However, as I continued modulating the worldview and updating details in Office Hours, I realized that different AI systems occupy unequal positions within the market, and these market advantages ultimately feed back into the company that creates and continuously refines them. AI appears neutral and trustworthy, but once the capitalists behind these systems begin using them for profit, the manipulation becomes much more difficult for ordinary people to detect.
Based on AI’s tendency to encourage trust, I decided to exaggerate its capabilities in my IP project. In the project, a company called Everyday Holiday, which specializes in selling holiday gifts and greeting cards, launches an AI assistant called DION. The company advertises DION as a highly accurate AI that has read vast amounts of authentic historical documents and can provide “real historical records” explaining the origins of holidays. Because there are relatively few major holidays in June, related industries are forced into periods of economic stagnation. To stimulate consumer demand, Everyday Holiday falsely promotes June 9 (the date of my first final exam in this quarter) as a revived version of the ancient “Bacchanalia”. The company name was something I happened to adopt while using ChatGPT to generate sample promotional pages for the company. I felt the name well captured the spirit of the project—every day is a festival; every day the company turns a profit. The connection between the AI character DINO and Bacchanalia runs even deeper. I used different variations of “Dionysus” to shape the close bond between them. After establishing Everyday Holiday and DION as the core of my project, I began searching for the most appropriate writing medium.
At first, I believed that a blog format would work best because it could simulate authentic public reactions. A blog would allow the project to present the spread of misinformation from the perspective of ordinary people in the future, demonstrating both the scale of the false information’s circulation and the negative cultural consequences produced when misinformation becomes embedded in collective consciousness. However, I eventually realized that this format made it difficult to emphasize AI’s role in the process. It was hard to distinguish whether the misinformation “came from AI” or “came from a holiday celebration company.” The latter felt much closer to nowadays life, but it also weakened the sense of defamiliarization that I wanted to create. As I continued gathering materials and building the website on GitHub and Netlify, I realized that designing a fictional corporate website would better support my concept. I drew inspiration from the homepage layouts of Hallmark, High Country News, and Apple. This format allows me to use astroturfing-style commercial propaganda to preserve my original idea of misinformation becoming embedded within collective consciousness, while simultaneously demonstrating DION’s influence through advertisements and promotional language.
For the writing section, I drew reference from the promotional pages of companies that sell festive goods, as well as discussions on social media about the social issues brought about by festivals. (Trammell, Montclair) After reading several articles from CNN and Substack, I began writing in the tone of a news broadcast. Once the first draft was complete, I noticed some areas that lacked careful consideration—in order to portray contrasting attitudes, my word choices were quite forceful and unsuitable for publication on a public platform. For instance, in the first half, when addressing the friend who raised doubts, I used absolute negating statements. After revision, my news structure was divided into a clear framework: everyone is celebrating Bacchanalia, and a minority opposes it because they don’t believe in the “facts”. This aligns with the core content of my project but lacks authenticity. I was reminded of real-world news broadcasts, where reporters offer precise analysis of ongoing events. When a social phenomenon suddenly rises to prominence in the future, there will inevitably be those who exaggerate its significance to the point of exhausting others. If my news piece could step in to counter these overly inflated expressions of value, it would lend the piece a sense of objectivity. Under this veneer of objectivity, the bigger question — does Bacchanalia exist—can be quietly overlooked.
Zewe, Adam. "A Better Method for Identifying Overconfident Large Language Models." MIT News, Massachusetts Institute of Technology, 19 Mar. 2026, news.mit.edu/2026/better-method-identifying-overconfident-large-language-models-0319.
Kush. "AI Governance." IBM AI Academy, IBM, 31 Oct. 2023, www.ibm.com/think/videos/ai-academy/generative-ai-governance..
Trammell, Kendall. "Mother's Day Founder Regretted Inventing the Holiday." CNN, 13 May 2018, www.cnn.com/2018/05/13/us/mothers-day-anna-jarvis-trnd/index.html.
Montclair State University. "Study: Valentine's Day 2023 Is All About Gifts, Less About Love." Montclair State University News Center, Montclair State University, 14 Feb. 2023, www.montclair.edu/newscenter/2023/02/14/valentines-day-2023-is-all-about-gifts-less-about-love-research-study/.