Digital society | Artificial intelligence

Welcome to Content Section 3.6: Artificial Intelligence (AI)

Hello Digital Society students! This chapter explores one of the most transformative technologies of our time: Artificial Intelligence (AI). AI is no longer science fiction—it influences everything from the music you stream to the ads you see and the decisions made by governments and corporations.
Understanding AI is vital because it directly relates to key course concepts like Power, Values and ethics, and Change. Don't worry if the technical aspects seem tricky; we will focus on the social, economic, and political implications, which is the core focus of Digital Society!

1. Defining Artificial Intelligence (AI)

What is AI? A Simple Definition

AI refers to computer systems designed to perform tasks that typically require human intelligence. These tasks include learning, problem-solving, decision-making, speech recognition, and visual perception.

Key Term: Artificial Intelligence (AI)
Systems, powered by algorithms (3.2), that simulate human cognitive processes to make predictions, solve complex problems, or take actions based on input data (3.1).

Analogy: The Smart Appliance

Think of your refrigerator. A regular fridge just chills things. An AI-enabled smart fridge might learn your grocery habits, predict when you're running out of milk, and order it for you. It performs a task (ordering) that requires a low level of human-like decision-making (assessing stock levels and timing).

2. The Different Types of AI

To analyze AI's impact, it’s essential to know that not all AI is created equal. We usually classify AI based on its capability—how "smart" it is compared to a human. The syllabus asks you to Identify different types of existing AI, so pay close attention to the distinctions below.

Type 1: Artificial Narrow Intelligence (ANI) / Weak AI

This is the only type of AI that currently exists and is widely deployed.

• Definition: ANI is designed and trained to perform a single, specific task exceptionally well. It operates within a pre-defined range.
• Capability: ANI cannot perform tasks outside of its programming. It mimics intelligence but doesn't truly understand the context or have consciousness.
• Real-World Examples:
  • Voice Assistants (Siri, Alexa)
  • Spam Filters in your email
  • Recommendation Systems (Netflix, Spotify)
  • Facial Recognition software
  • Self-driving car systems (focused only on driving).
• Key Takeaway: If an AI system is excellent at one thing but useless at everything else, it is Narrow AI.

Type 2: Artificial General Intelligence (AGI) / Strong AI

This type of AI currently only exists in theory and fiction.

• Definition: AGI would possess the ability to understand, learn, and apply intelligence to solve any problem that a human being can.
• Capability: It would have consciousness, self-awareness, and the ability to transfer learning from one domain to an entirely new one (e.g., learning how to drive and immediately applying that abstract knowledge to piloting a drone).
• Real-World Examples: Think C-3PO or Data from Star Trek. We are not there yet!

Type 3: Artificial Superintelligence (ASI)

This is purely hypothetical and highly speculative.

• Definition: ASI would be a system where intelligence exceeds human cognitive capability in virtually every aspect—scientific creativity, general knowledge, and social skills.
• Implication: This level of AI would drive unprecedented Change and raises huge questions about Power and human control.

3. How AI Works: The Role of Data and Algorithms

AI relies heavily on two foundational elements of the Digital Society content: Data (3.1) and Algorithms (3.2).

The Engine of AI: Machine Learning (ML)

The most common way to create AI today is through Machine Learning (ML). ML is a method where algorithms are trained on vast amounts of data, allowing the system to improve its performance without being explicitly programmed for every possible scenario.

Step-by-Step: Training an ML System (The Puppy Analogy)

1. Gather Data: Collect thousands of examples (e.g., pictures of cats and dogs, labeled correctly). This is the "Data Input" (3.1).
2. Feed the Algorithm: The ML algorithm (3.2) analyzes the data, looking for patterns and correlations (e.g., dogs have longer noses, cats have pointed ears).
3. Prediction and Feedback: The system makes a prediction (Is this a cat or a dog?). If it’s wrong, the algorithm adjusts its internal weightings based on the error. This is the learning part.
4. Refinement: Repeat steps 2 and 3 millions of times until the algorithm is highly accurate. It has now "learned" how to distinguish between the two animals.

Did you know? The performance of an AI system is almost entirely dependent on the quality and quantity of the data it is trained on.

4. Impacts and Implications of AI in Digital Society

In the IB Digital Society course, our primary focus is the impact AI has on people, communities, and contexts (cultural, economic, social, political). AI systems often intersect with the concept of Values and ethics (2.7).

A. Ethical and Social Implications

1. Algorithmic Bias and Discrimination

This is perhaps the most serious ethical concern. If the data used to train an AI reflects existing societal inequalities (such as racial or gender bias), the resulting AI will learn and perpetuate those biases, often making them worse.

• Impact: AI used in judicial systems might disproportionately recommend harsher sentences for certain demographics based on biased historical crime data. AI loan approval systems might unfairly deny credit based on zip codes or names.
• Concept Link: This directly challenges Values and ethics (2.7) and raises questions about Equity and Fairness.

2. Transparency and the "Black Box" Problem

Many advanced AI systems (especially deep learning models) are so complex that even their creators cannot fully explain why they reached a specific decision. This is called the Black Box Problem.

• Impact: If an AI wrongly rejects a job application or misdiagnoses a patient, it is incredibly difficult to challenge the decision or hold the system accountable. This lack of accountability affects Governance and human rights (HL Extension).

B. Economic and Political Implications

1. Automation and Employment (Economic Context)

AI drives significant Change (2.1) by automating tasks that were previously done only by humans.

• Routine Tasks: AI is excellent at automating repetitive tasks (e.g., data entry, customer service chatbots).
• The Impact: While automation can increase efficiency and productivity (good for the economy), it can lead to widespread job displacement, requiring huge societal investments in retraining and education.
• Example: Self-checkout kiosks, automated stock trading.

2. Manipulation and Political Power (Political Context)

AI systems are powerful tools for targeting, persuasion, and surveillance.

• Targeted Information: Political campaigns use AI to analyze voter data and deliver highly specific, personalized messages, which can increase polarization and spread misinformation (affecting Expression (2.2)).
• Surveillance: Governments increasingly use AI-powered facial recognition and monitoring tools, raising concerns about privacy, individual freedoms, and state Power (2.4).