====== What is Artificial Intelligence ======

**Artificial Intelligence (AI)** is the simulation of human intelligence processes by computer systems, encompassing learning, reasoning, problem-solving, perception, and language understanding.((IBM. "Understanding the different types of artificial intelligence." [[https://www.ibm.com/think/topics/artificial-intelligence-types|IBM Think]])) Since its formal inception in the 1950s, AI has evolved from a theoretical curiosity into a foundational technology reshaping economies, institutions, and everyday life across the globe.

As of 2026, AI is no longer treated as a future technology. It has become infrastructure — always present, deeply embedded, and increasingly expected across virtually every industry and domain of human activity.((Armani, Sydney. "The State of AI in 2026: The Year Intelligence Became Infrastructure." [[https://aiworldjournal.com/the-state-of-ai-in-2026-the-year-intelligence-became-infrastructure/|AI World Journal]], February 2026.))

===== Definition and Core Concepts =====

At its core, artificial intelligence refers to machines that can perform tasks typically requiring human intelligence. These tasks include:

  * **Learning** — acquiring information and rules for using it
  * **Reasoning** — using rules to reach approximate or definite conclusions
  * **Problem-solving** — finding solutions to complex challenges
  * **Perception** — interpreting sensory input such as images, speech, or text
  * **Language understanding** — processing and generating human language

The field draws from computer science, mathematics, linguistics, psychology, neuroscience, and philosophy. Modern AI systems use statistical methods and vast datasets to identify patterns, make predictions, and generate new content.((Swiss Cyber Institute. "History of Artificial Intelligence." [[https://swisscyberinstitute.com/blog/history-artificial-intelligence/|Swiss Cyber Institute]]))

===== History of Artificial Intelligence =====

The history of AI spans more than 80 years, marked by periods of intense optimism, funding booms, disappointing "AI winters," and transformative breakthroughs.

==== Early Foundations (1940s-1950s) ====

The intellectual foundations of AI trace to the 1940s. In 1942, Alan Turing built the Bombe machine to crack Enigma codes during World War II, demonstrating that machines could solve problems at speeds far beyond human capability.((Big Human. "History of Artificial Intelligence." [[https://www.bighuman.com/blog/history-of-artificial-intelligence|Big Human]])) In 1950, Turing published his landmark paper "Computing Machinery and Intelligence," proposing the **Turing Test** — a method to evaluate whether a machine can exhibit intelligent behavior indistinguishable from a human. That same decade, Arthur Samuel created the first self-learning checkers program in 1952, introducing core machine learning concepts.

==== The Dartmouth Conference and Birth of AI (1955-1960s) ====

In 1955, **John McCarthy** — later known as the "Father of AI" — coined the term "artificial intelligence" in a proposal for a workshop at Dartmouth College. The **Dartmouth Conference of 1956** officially launched AI as an academic field, bringing together McCarthy, Marvin Minsky, Nathaniel Rochester, and Claude Shannon. McCarthy subsequently developed the LISP programming language in 1958, which became the dominant language for AI research for decades.

Allen Newell, Herbert Simon, and Cliff Shaw created the **Logic Theorist** in 1955-1956, the first program capable of proving mathematical theorems — widely considered the first true AI program.

==== First AI Winter and Expert Systems (1970s-1980s) ====

The initial optimism of the 1960s gave way to disappointment as early AI systems failed to deliver on ambitious promises. The **first AI winter** struck in the 1970s as government funding dried up due to computational limitations and overhyped results.

The 1980s saw a resurgence through **expert systems** — rule-based programs designed to mimic human specialists in narrow domains like medical diagnosis and financial analysis. However, these systems proved brittle and expensive to maintain, leading to a **second AI winter** in the late 1980s and early 1990s.

==== Resurgence and Key Milestones (1990s-2010s) ====

AI experienced a gradual revival driven by increasing computational power and data availability:

  * **1997** — IBM's **Deep Blue** defeated chess world champion Garry Kasparov
  * **2011** — IBM's **Watson** won Jeopardy!, demonstrating data-driven AI capabilities
  * **2012** — The **AlexNet** deep learning breakthrough in image recognition launched the deep learning revolution
  * **2016** — Google DeepMind's **AlphaGo** defeated Go world champion Lee Sedol, showcasing neural networks for complex strategic planning

==== The Deep Learning and Generative AI Era (2017-Present) ====

The introduction of the **Transformer architecture** in 2017 proved to be a watershed moment, enabling the modern generation of language models.((SalesApe. "An AI Timeline: 2020-2025." [[https://www.salesape.ai/articles/an-ai-timeline-2020-2025|SalesApe]])) Key milestones include:

  * **2020** — DeepMind's **AlphaFold** solved the protein folding problem
  * **2020-2021** — OpenAI released **GPT-3** and **DALL-E**, introducing human-like text generation and multimodal image generation
  * **2022** — **ChatGPT** launched publicly, sparking a global generative AI boom; GitHub Copilot brought AI-assisted coding to developers
  * **2023-2024** — Rapid advances in multimodal models, with GPT-4, Claude 3, and Gemini pushing capabilities further
  * **2025** — Context windows expanded to 10 million tokens; humanoid robots reached consumer price points; reasoning models moved from premium features to free-tier defaults((Khalaf, Farida. "2025 AI Year in Review." [[https://medium.com/@Fofa/2025-ai-year-in-review-the-complete-roadmap-of-breakthroughs-models-infrastructure-46b32e456e67|Medium]], February 2026.))
  * **2026** — The rise of **agentic AI** — systems capable of executing multi-step tasks with minimal supervision, marking the shift from AI as a tool to AI as infrastructure

===== Types of Artificial Intelligence =====

AI is commonly classified by **capability level**:

^ Type ^ Description ^ Current Status ^
| **Artificial Narrow Intelligence (ANI)** | Designed for specific tasks; cannot generalize beyond its training | Widely deployed; powers all current commercial AI |
| **Artificial General Intelligence (AGI)** | Would match human-level reasoning across diverse tasks | Theoretical; no true AGI exists, though advanced models approach narrow benchmarks |
| **Artificial Super Intelligence (ASI)** | Would surpass human intelligence in all aspects | Purely speculative; no development milestones achieved |

All AI systems in production as of 2026 are forms of **Narrow AI**, though modern multimodal models are significantly more capable than earlier narrow systems. Some researchers use terms like "proto-AGI" or "agentic AI" to describe systems that exhibit cross-domain capabilities while still being bounded by their training.((Wang, Alex. "The 7 Types of AI — Explained (with 2025 Reality Check)." [[https://www.linkedin.com/pulse/7-types-ai-explained-2025-reality-check-alex-wang-ybktc|LinkedIn]], November 2025.))

===== Major Subfields of AI =====

==== Machine Learning (ML) ====

Machine learning is the subfield of AI focused on algorithms that learn from data without being explicitly programmed. It encompasses supervised learning (training on labeled data), unsupervised learning (finding patterns in unlabeled data), and reinforcement learning (learning through trial and error with rewards). ML is the foundation for most modern AI applications.

==== Natural Language Processing (NLP) ====

NLP enables machines to understand, interpret, and generate human language. Advanced by the Transformer architecture and models like GPT and Claude, NLP powers chatbots, translation services, content generation, and document analysis.

==== Computer Vision ====

Computer vision allows machines to interpret and analyze visual information from images and video. Revolutionized by the 2012 deep learning breakthrough, it enables facial recognition, medical imaging analysis, autonomous vehicle navigation, and quality control in manufacturing.

==== Robotics ====

Robotics integrates AI with physical machines to perform tasks in the real world. In 2025-2026, humanoid robots reached consumer price points (approximately $5,900), and companies like Amazon began training robots for package delivery and warehouse operations.((Wikipedia. "Timeline of artificial intelligence." [[https://en.wikipedia.org/wiki/Timeline_of_artificial_intelligence|Wikipedia]]))

===== Current State of AI (2025-2026) =====

The AI landscape in 2025-2026 is defined by several key trends:

**Agentic AI:** The most significant shift is the rise of AI agents — systems capable of planning, deciding, and acting across multi-step tasks with minimal human supervision. These agents can perform competitive research, generate marketing campaigns, manage customer support workflows, run financial forecasting, and automate operations across departments.

**Multimodal Models:** Leading AI systems now process text, images, audio, and video natively. Context windows have expanded dramatically — from 128,000 tokens to 10 million tokens in a single year.

**Infrastructure-Scale Deployment:** AI spending hit $61 billion in infrastructure alone in 2025. Approximately 71% of organizations regularly use generative AI, with 96% of enterprise IT leaders reporting some level of AI integration.((AmplifAI. "90+ Generative AI Statistics You Need to Know in 2026." [[https://www.amplifai.com/blog/generative-ai-statistics|AmplifAI]], March 2026.))

**Regulation:** The EU AI Act entered its enforcement phase with prohibitions taking effect in August 2024 and full applicability expected by August 2026. France's AI Action Summit in February 2025 saw 61 countries sign a declaration on AI governance.

===== Key Applications Across Industries =====

^ Industry ^ Key AI Applications ^
| **Healthcare** | Medical imaging diagnostics, drug discovery (AlphaFold), patient care optimization, disease progression simulation |
| **Finance** | Fraud detection, algorithmic trading, risk assessment, automated compliance |
| **Software Development** | Code generation (GitHub Copilot, Cursor), automated testing, debugging assistance |
| **Manufacturing** | Quality control, predictive maintenance, robotics automation, supply chain optimization |
| **Retail** | Recommendation engines, inventory management, self-checkout systems, demand forecasting |
| **Transportation** | Autonomous vehicles (450,000 driverless rides per week in 2025), route optimization, logistics planning |
| **Creative Industries** | Image and video generation, content creation, design assistance, music composition |
| **Education** | Personalized learning, automated grading, intelligent tutoring systems |
| **Legal** | Document review, contract analysis, legal research assistance |

===== See Also =====

  * [[ai_models|What is an AI Model]]
  * [[types_of_ai|Types of AI]]
  * [[generative_ai|Generative AI]]
  * [[ai_ethics|Ethical Concerns of AI]]
  * [[future_of_work_ai|How AI Will Impact the Future of Work]]
  * [[ai_providers_vs_models|AI Providers vs AI Models]]

===== References =====