# Intro

## What is AI?

### Acting Humanly

- The **Turing Test** 圖靈測試
- Total Turing Test - the computer will need:
- Computer vision 電腦視覺 to perceive objects.
- Robotics 機械部件 to manipulate objects and move about.
- Problem: Turing test is not reproducible（可再現的）, constructive（有建設性的）, or amenable（順從的） to mathematical analysis.

### Thinking Humanly

- **Cognitive Science** 認知科學
- How to validate? Requires
- Predicting and testing behavior of human subjects (top-down)
- Direction identification from neurological 神經學 data (bottom-up)

### Thinking Rationally

- **Laws of Thought** 思想法則
- Problems:
- Not all intelligent behavior is mediated by logical deliberation (並非所有智能行為都是通過邏輯深思熟慮來調節的)
- What is the purpose of thinking? What thoughts should I have out of all the thoughts (logical or otherwise) that I could have? (思考的目的是什麼？在我可能有的所有想法（邏輯或其他）中，我應該有什麼想法？)

### Acting Rationally

- Rational behavior - doing the right thing
- The right thing - that which is expected to maximize goal achievement, given the available information (鑑於現有信息，預計將最大限度地實現目標)

### Rational Agents

- An agent is an entity that perceives and acts
(代理是感知和行動的實體)
- Abstractly, an agent is a function from percept histories to actions:
- f: P* $\rightarrow$ A
- For any given class of environments and tasks, we seek the agent (or class of agents) with the best performance
- Caveat (警告): computational limitations make perfect rationality unachievable (計算限制使得完美理性無法實現)
$\rightarrow$ design best program for given machine resources
### AI Prehistory

- Philosophy 哲學
- Mathematics 數學
- Psychology 心理學
- Economics 經濟學
- Linguistics 語言學
- Neuroscience 神經科學
- Control theory 控制理論

### Risks and Benefits of AI

> "First solve AI, then use AI to solve everything else."
> [name=Demis Hassabis, CEO of Google DeepMind]

- Benefits
- Decrease repetitive work
- Increase production of goods and services
- Accelerate scientific research (disease cures, climate change and resource shortage solutions)
- Risks
- Lethal autonomous weapons 致命自主武器
- Surveillance and persuasion 監視和勸說
- Biased decision making 決策有偏見
- Impact on employment 就業的影響
- Safety-critical applications 安全關鍵性應用
- Cybersecurity threats 網路安全威脅
- Development of an artificial superintelligence that surpasses human intelligence may pose a significant risk
- Analogous to the "Gorilla problem"
- Humans and gorillas evolved from the same species, but humans have more control than other primates
- Thus, we should design AI systems in such a way that they do not end up taking control in the way Turing suggests they might