AI

Quote

Imagine what it can do for the process of innovation, for discovering new materials, medicine, energy, climate, and so many of the present challenges that we face as a species.” 

— Darío Gil, IBM

Definition

Simulation of human intelligence in machines, enabling them to perform tasks such as:

• Learning
• Reasoning
• Problem-solving
• Decision-making

Strength Levels

• Weak AI
• Strong AI
• Super AI

Subsets of AI

• Machine Learning
  • Deep Learning
  • Foundation Models
    • LLMs
• Generative AI
• Reinforcement Learning (e.g., game AIs)

History

• 1960s: Eliza mimics conversations.
• 1980s: Machine Learning boom.
• 1990s: Neural Networks.
• 2000s: Deep Learning.
• 2010s: NLP and computer vision.

Multidisciplinary Approach

• Engineering and Science: Robots and hardware like GPUs for training massive data models.
• Math and Statistics: Foundations of the models.
• Philosophy: Ethics.
• Politics: Applications and ethics.

Defining Innate Intelligence

Innate intelligence is tied to a purpose. For example, the intelligence of a seed is what makes it grow into an oak tree.

We have yet to define consciousness in speech, let alone in mathematics or code to create Super AI, so it will not emerge anytime soon.

Do We Truly Have “Artificial Intelligence”?

• Memorization: Databases have been around for a while.
• Calculation: Calculators can multiply big numbers.
• Mastery: DeepBlue has beaten the best human chess players.
• Turing Test: Involves a human chatting with both an AI and another human. If the human can’t distinguish between them, the AI passes the test.

AI in Daily Life

• Personalizing Experiences:
  • Netflix recommends shows based on your preferences.
  • Facebook and Instagram personalize content.
• Streamlining Tasks with Automation: AI improves day-to-day convenience.
• Personal Assistants:
  • Speech-based like Siri and Alexa, or text-based like ChatGPT.
  • Customer service chatbots.
  • Medication reminders.
• Internet of Things (IoT): Processes real-time data and automates tasks like home temperature control.
• Security: Enhances biometric recognition and detects financial fraud.
• AI-Powered Wearables: Monitors heart rate, oxygen levels, and other metrics during sleep.
• Smartphone Cameras: Features facial recognition, scene detection, portrait mode, etc.
• Image Editing: Removes backgrounds, silhouettes, etc.
• Autocorrect: Improves text accuracy.

Artificial vs Augmented Intelligence

• Augmented Intelligence:
  • Enhances human abilities (e.g., collision detection, screen readers).
  • Machines and humans work together to complement each other’s efforts.
• Artificial Intelligence:
  • Machines perform tasks requiring human intelligence (e.g., reasoning, problem-solving).
  • Replaces the need for a human.

Machines	Humans
Ingesting Data	Generalizing
Repetitive Tasks	Creativity
Accurate	Emotional Intelligence

Types of Learning

• Supervised Learning
• Unsupervised Learning
• Reinforcement Learning
• Self-Supervised Learning

Types of AI

1. Diagnostic/descriptive AI

   Focuses on assessing the correctness of behavior by analyzing historical data to understand what happened and why. This type of AI is instrumental in identifying patterns and trends, performing comparative analyses, and conducting root cause analyses.

   Capabilities:

   • Scenario planning: Helps in creating different future scenarios based on historical data.
   • *Pattern/trends recognition: Identifies recurring patterns and trends within data sets.
   • *Comparative analysis: Compares various data points to find correlations and insights.
   • *Root cause analysis: Determines the underlying reasons behind specific outcomes.

2. Predictive AI

   Forecasting future outcomes based on historical and current data. This type of AI is used extensively in predicting customer behavior, market trends, and other forward-looking insights.

   Capabilities:

   • *Forecasting: Predicts future trends and events.
   • *Clustering and classification: Groups similar data points and classifies them into predefined categories.
   • *Propensity model: Assesses the likelihood of specific outcomes based on current data.
   • *Decision trees: Utilize a tree-like model of decisions to predict outcomes.

3. Prescriptive AI

   Prescriptive AI focuses on determining the optimal course of action by providing recommendations based on data analysis. It goes beyond prediction by suggesting actions that can help achieve desired outcomes.

   Capabilities:

   • *Personalization: Tailors recommendations and experiences to individual needs.
   • Optimization: Identifies the most efficient ways to achieve goals.
   • Fraud prevention: Detects and prevents fraudulent activities through analysis.
   • *Next best action recommendation: Provides actionable insights on the next steps to take.

4. Generative/cognitive AI

   Generative or cognitive AI is involved in producing various types of content, such as code, articles, images, and more. This type of AI mimics human creativity and cognitive processes to automate and assist in content creation.

   Capabilities:

   • *Advises: Offers expert advice and recommendations.
   • *Creates: Produces new content, such as text, images, and code.
   • *Protects: Enhances security measures through intelligent analysis.
   • *Assists: Provides assistance in various tasks, improving efficiency.
   • *Automates: Automates repetitive tasks to save time and resources.

5. Reactive AI

   Reactive AI systems are designed to respond to specific inputs with predetermined responses. They do not have memory or the ability to learn from past experiences, making them suitable for tasks that require immediate reactions.

   Capabilities:

   • *Rule-based actions: Executes specific actions based on predefined rules.
   • Instant responses: Provides immediate responses to inputs.
   • *Static data analysis: Analyzes current data without considering past interactions.

   Example:

   • An OpenSea bot that takes an offer for an NFT if it exceeds a predetermined amount.

6. Limited memory AI:

   Limited memory AI systems have the ability to use past experiences to inform current decisions. They can learn from historical data to improve their performance over time. This type of AI is commonly used in autonomous vehicles and recommendation systems.

   Capabilities:

   • *Learning from data: Uses historical data to make informed decisions.
   • *Pattern recognition: Identifies patterns over time to improve accuracy.
   • Adaptive responses:** Adapts responses based on previous interactions.

7. Theory of Mind AI:

   Theory of Mind AI is an advanced type of AI that aims to understand human emotions, beliefs, and intentions. It is still in the research stage and seeks to interact more naturally with humans by comprehending their mental states.

   Capabilities:

   • *Emotion recognition: Identifies and responds to human emotions.
   • *Social interaction: Engages in more natural and human-like interactions.
   • *Intent prediction: Predicts human intentions based on context and behavior.

8. Self-aware AI:

   Self-aware AI represents the most advanced form of AI, which has its own consciousness and self-awareness. This type of AI can understand and react to its own emotions and states. It remains a theoretical concept and has not yet been realized.

   Capabilities:

   • *Self-diagnosis: Evaluates its own performance and health.
   • *Autonomous learning: Learns independently without human intervention.
   • Adaptive behavior:* Adjusts behavior based on self-awareness.

9. Narrow AI (Weak AI:

10. General AI (Strong AI)

Future of AI

It is not about one model, about ChatGPT or Deepseek, it is about multiple models applied. The