What is Artificial Intelligence (AI)

Artificial Intelligence (AI) is the field of building systems that perform tasks that normally require human intelligence — reasoning, perception, language understanding, planning, and decision making under uncertainty.

The birth of AI (1956 Dartmouth Conference)

AI became a formal research discipline at the Dartmouth Summer Research Project in 1956, organized by John McCarthy with pioneers including Marvin Minsky, Nathaniel Rochester, and Claude Shannon. Their working hypothesis: every aspect of learning or intelligence can be precisely described so that a machine can simulate it.

Behavioral intelligence: the Turing Test (1950)

Alan Turing proposed the Imitation Game: a human judge chats via text with a human and a machine. If the judge cannot reliably identify the machine, the system demonstrates behavioral intelligence. Early systems like ELIZA (1966) fooled users with pattern matching and reflection — not true understanding, but a useful benchmark for conversational interfaces.

The four quadrants of AI

AI is often classified along two axes — process vs. behavior and human-like vs. rational:

Thinking humanly — cognitive modeling of human thought structures.
Acting humanly — behavior judged by human-relative criteria (e.g. passing the Turing Test).
Thinking rationally — logic and formal inference (syllogisms, theorem proving).
Acting rationally — agents that choose actions to maximize expected utility from data and constraints.

Modern enterprise AI is overwhelmingly aligned with acting rationally: minimize objective loss, hit task goals reliably, rather than mimic human cognitive quirks.

Classical AI (GOFAI) and expert systems

Good Old-Fashioned AI (GOFAI) relied on symbolic, hand-built knowledge:

Formal logic — facts + inference rules (A → B, B → C ⇒ A → C).
Production rules — IF condition THEN action trees.
Semantic networks — entities as nodes, relations as edges (e.g. Sparrow → is-a → Bird).

Expert systems (1970s–80s) split a knowledge base (domain rules from experts) from an inference engine that applied them. MYCIN (1975) diagnosed blood infections and recommended antibiotics — a landmark commercial symbolic system.

Why GOFAI failed and AI winters followed

Ambiguity — brittle translation and context failures outside toy domains.
Combinatorial explosion — rule sets grew faster than systems could search them.
AI winters — funding collapses in the mid-1970s and late 1980s when symbolic AI could not scale to messy real-world data.

flowchart LR A[Symbolic AI / GOFAI] --> B[Expert Systems] B --> C[AI Winters] C --> D[Machine Learning] D --> E[Deep Learning] E --> F[Generative AI]

Real-world examples today

Route optimization and fraud detection in production systems.
Spam filtering and recommendation engines.
Voice assistants, coding copilots, and enterprise chatbots.