Intro to AI: History and Definition

• AI is defined as augmented intelligence
• AI aims to extend human capabilities and tackle tasks beyond human and machine capabilities.

History

• AI's history traces back to the origins of computing with early automation efforts.
• 1950s: Alan Turing proposed the Turing Test, and John McCarthy coined "artificial intelligence."
• 1960s: Early programs such as ELISA and SHRDLU were created.
• 1970s: Expert systems rose to prominence.
• 1980s: Machine learning surged, setting the stage for future progress.
• 1990s: Neural networks were introduced.
• 2000s: Deep learning ascended.
• 2010-2020: AI applications spread across industries, including NLP and computer vision.
• Present Decade: Rapid expansion of AI continues with advancements in deep learning, autonomous systems, and healthcare.

Definition of AI

• AI is the simulation of human intelligence processes by computer systems.
• AI uses algorithms and data to enable machines to learn, reason, problem-solve, and make decisions.
• AI ranges from simple automation to complex deep learning and neural networks.

Enablers of AI

• The internet has revolutionized connectivity, providing faster access to more information.
• Distributed computing scales data processing, enabling efficiency.
• IoT proliferation of connected devices generates massive data sets.
• Social networking has encouraged data to be unstructured.
• Together, they reshape our digital landscape, accelerating information access and innovation.
• Augmented intelligence places information at experts' fingertips, backed by evidence, for informed decisions.

Innate vs. AI Intelligence

• Humans possess innate intelligence governing bodily functions.
• Machines' innate intelligence is derived from human programming.
• Machines learn by examining examples and creating machine learning models based on inputs and desired outputs.
• Machine learning models are developed through supervised learning, unsupervised learning, and reinforcement learning.

Types of AI based on Strength

• AI can be divided into categories based on strength, breadth, and application.

Weak or Narrow AI

• AI applied to a specific domain.
• Performs specific tasks but cannot learn new ones.
• Makes decisions based on programmed algorithms and training data.
• Examples: language translators, virtual assistants, AI-powered web searches, recommendation engines, and intelligent spam filters.

Strong or Generalized AI

• AI capable of performing diverse and unrelated tasks.
• Can acquire new skills to tackle novel challenges by autonomously learning fresh approaches.
• Combination of many AI strategies that learn from experience and can perform at a human level of intelligence.

Super or Conscious AI

• Advanced concept extending generative AI.
• AI with human-level consciousness, self-awareness, advanced cognitive abilities, and independent thinking.
• Unlikely to be created in the near future due to the challenge of defining consciousness.
• Could demonstrate capabilities beyond human intelligence in areas like healthcare, autonomous vehicles, robotics, natural language understanding, and environmental conservation.

Fields Contributing to AI

• Computer science and electrical engineering: Determine how AI is implemented in software and hardware.
• Mathematics and statistics: Determine viable models and measure performance.
• Psychology and linguistics: Essential in understanding how AI might work, based on brain function.
• Philosophy: Provides guidance on intelligence and ethical considerations.