Week1

When you hear the words “Artificial Intelligence”, what are the first four things that come to your mind?

1. Algorithms and Automation

AI is fundamentally driven by algorithms and the development of computer technologies. Its creation involves supplying data, training models, and moving from machine learning into deep learning. Through this process, AI acquires automated capabilities, which means it can complete assigned tasks or even perform work with minimal human intervention.

2. Applications and the Future

AI is already integrated into many aspects of our everyday lives, such as chatbots, recommendation systems, and smart home devices. These applications bring convenience and efficiency, but they also spark both excitement and concern about the future. On one hand, AI can support creativity, healthcare, and even education, on the other, it also raises questions about privacy and over-reliance on technology.

3. Human–AI Relationships and Ethics

Because AI models are designed, trained, and shaped by humans, they inevitably reflect human choices, biases, and values. AI systems can surpass human capabilities in speed and scale of analysis, which reshapes the relationship between humans and machines. This dynamic creates both opportunities (such as new job roles) and risks (such as widespread job displacement). It also demands careful attention to responsibility, accountability, and clear boundaries in how AI is developed and applied.

Think about the devices and/or digital services you use daily. Write below a list of the top three that are present in your life. (p.23)

Laptop: Studying, completing assignments, and working on creative projects.

Digital Maps:  Navigation and commute to any new places.

Music Platforms: Entertainment and explore new music genres and artists for inspiration of my projects.

Have these things ever surprised you by guessing something about you that you didn’t expect? Here are some examples:

“One morning I noticed my phone created a heartfelt recap of the year with my photos that made me tear up a little.”

“One time I got an email and my email inbox suggested how I should reply to it.”

See if you can recall a similar moment and write about it below. (p.24)

One time I was browsing an online shopping platform and noticed that it recommended an item I had only recently been thinking about but hadn’t searched for yet, as if the system could read my mind. On the one hand, I was impressed by how convenient and personalized the recommendation was. On the other hand, it also made me wonder how much data the platform had collected about me, and whether I was really comfortable with that level of prediction.

AI is very present in smartphones and Internet-based applications we use everyday. Sometimes, it might feel strange when you notice an application of AI—it might feel like someone is listening in on you and keeping track of your habits. Other times, you might think that it’s useful, or helpful for something to know your patterns. Either way, being able to identify the many roles that AI can play can help you begin to think about how it impacts your daily life.

Take a moment to see if you can identify what function AI plays in the following list. If you get lost, go back to the examples of AI In Action on page 17. (p. 25)

• Device / Service: AI Function(s)

• Email inbox: Spam filtering, auto-sorting of important emails, smart reply suggestions

• Check depositing: Image recognition to scan checks, scam detection

• Texting & mobile keyboards: Autocorrect, predictive text suggestions, emoji recommendations

• Netflix: Personalized content recommendations

• Google (search function): Search ranking algorithms, autocomplete suggestions

• Social media platforms (Instagram, Facebook, Twitter/X, etc.): Content recommendation feeds, facial recognition in photos, targeted ads

• Automated message systems: Natural language processing to understand and respond to user queries

Now that you have a better idea of what applications of AI in everyday life look like, let’s think about the impact that they have.

What do we gain by having AI in our everyday lives?(p.26)

AI brings a lot of convenience to our daily life. Intelligent search tools save a lot of time and make it easier to find information quickly. And recommendation systems suggest personalized products or content that we might actually like. Moreover, AI-powered services often help us solve problems more efficiently, which makes our life feel easier.

What do we lose by having AI in our daily lives?(p.26)

At the same time, AI raises concerns about privacy, since so much of our personal data is collected and analyzed. It may also replace certain jobs, creating insecurity in the workplace. Finally, relying too much on AI can reduce human-to-human interaction, which makes me wonder how it might affect our social connections in the long run.

Think about what you have learned so far about what AI is and how it shows up in our daily lives. Your task is to identify a problem that you see in your life, neighborhood, or community and design an AI system that could help address this problem.(p.27)

What Problem to address:

The overcrowding in public transportation during rush hours: Subways and buses often become extremely packed, which leads to long waiting times, discomfort for passengers, and sometimes even safety concerns. This not only reduces efficiency but also creates stress for daily commuters.

How AI Helps:

AI could help by predicting passenger flow patterns based on historical and real-time data. An AI system could analyze usage trends to forecast which lines or stations will become crowded at certain times. It could also provide real-time updates about which train cars or buses have fewer passengers, so people can distribute more evenly. Additionally, AI could suggest alternative routes or recommend slight adjustments to departure times, encouraging commuters to travel in a more staggered way.

Role of Humans:

Commuters would make their final decisions about how and when to travel. AI would act just as a supportive tool that provides useful insights. Transportation authorities would also play an important role by responding to AI’s recommendations. In this way, AI does not replace human decision-making but make some suggestions.

Data Needed:

The system would need access to data such as subway entry and exit records, passenger numbers in per train car, and bus ridership levels. With this information, AI could create reliable predictions of crowd levels across different times and locations.

Responsible Data:

To protect privacy, all passenger data should be anonymized and stripped of personally identifiable information before analysis. Instead of tracking individual movements, the system would only analyze aggregated patterns. Data collection should also be transparent, with clear explanations to the public about what information is gathered, how it will be used, and how privacy is safeguarded. This ensures consent, trust, and responsible use of technology.

What the System Will Look Like? (p.28)

At the backend: a cloud-based AI platform that processes live sensor data and historical ridership records.

At the frontend: Digital dashboards for transportation authorities (maps with color-coded congestion levels, predicted flows).

Mobile app notifications for commuters.

In-station display screens that update in real time.

Where It Will Live?(p.28)

Physical Environment: Inside subway stations, buses, and train cars (sensors, cameras, display boards).

Digital Environment: On commuters’ phones (apps), on station announcement boards, and in transport authority control rooms.

Cloud / Data Center: The AI models live on secure servers where data is stored and analyzed.

I found that the model is already quite advanced and can correctly recognize most of the images I tested. Even when the confidence level was not very high, it still managed to identify the main subject in many cases. For the few images it failed to classify accurately, I think the problem was not with the model itself but rather with the complexity of the image. When the scene was visually complicated, the accuracy dropped noticeably. I also observed that the position and lighting of the subject had a clear impact. For example, when I uploaded an image where the main object was located near the bottom of the frame and the lighting was dim, the model struggled to give an accurate prediction. Overall, I believe the model is quite mature, but it still relies heavily on surface-level features learned from data rather than a deeper understanding of what the object truly is.