Free tools Windows power users keep installed
One-click scans. No signup required.
Charles Babbage designed a programmable mechanical computer, Ada Lovelace explained why such a machine could manipulate symbols beyond arithmetic, and Alan Turing formalized general-purpose computation before asking how machine intelligence might be recognized. They did not build modern artificial intelligence, but their ideas form part of its intellectual prehistory.
Before artificial intelligence: two very different Babbage engines
Babbage’s work began with automated calculation, not machine thought. In 1822 he presented the Difference Engine, a specialized mechanical calculator intended to produce mathematical tables while reducing human error. A portion was built, and later reconstructions showed that the design was mechanically workable. It was not a general-purpose programmable computer. The Science Museum describes the Difference Engine and its development.
During the 1830s Babbage developed the far more ambitious Analytical Engine. It was a design for a programmable mechanical machine using punched cards, partly inspired by the Jacquard loom. The engine was never completed in his lifetime, so calling it a functioning nineteenth-century computer is misleading. It is more accurate to call it one of the earliest detailed designs for a general-purpose programmable computer.
| Analytical Engine concept | Modern analogy |
|---|---|
| Mill | Processor or arithmetic unit |
| Store | Memory |
| Punched cards | Program instructions |
| Input and output mechanisms | Data entry and results |
These are useful analogies, not claims that Babbage had electronic components. The crucial change was conceptual: the machine was intended to follow different sets of instructions rather than perform one fixed calculation.
#1 Best Overall
Ada Lovelace sees beyond arithmetic
In 1843 Ada Lovelace translated Luigi Menabrea’s French account of the Analytical Engine and added notes substantially longer than the original article. Her notes explained the machine’s architecture, supplied examples, and explored what programmable symbolic operations might mean. The Computer History Museum provides background on the Babbage engines and Lovelace’s contribution.
The Bernoulli-number procedure
Note G described a method for calculating Bernoulli numbers. It is widely regarded as the first published computer program, or the first published program intended for a general-purpose machine. The absolute claim that Lovelace was the first person ever to write a program is too strong: Babbage had produced earlier unpublished program sketches. “First published computer programmer” is the safer historical description. Archival context is available from the Babbage Archive Project.
Why symbols mattered
Lovelace’s deeper insight was that numbers could represent things other than quantities. If musical notes, letters, or other formal objects were encoded according to rules, the engine could manipulate their representations. She argued that the Analytical Engine might, in principle, compose or transform music when supplied with the relevant data and instructions. This was an account of symbolic computation, not a description of consciousness or modern machine learning. The Science Museum explains this broader view of the machine.
Rank #2
Lovelace’s objection: programming is not automatically intelligence
Lovelace is often presented as a straightforward prophet of artificial intelligence. Her own writing is more complicated. She stated that “The Analytical Engine has no pretensions to originate anything” and would do whatever people knew how to order it to perform. In context, this was a claim about the designed mechanical engine and the source of its operations, not a settled verdict on every future form of AI. The passage appears in the discussion reproduced with Turing’s later response in Turing’s 1950 paper.
Her position separates two questions that are still confused today: how expressive a program can be, and whether a machine’s output counts as originality. Lovelace anticipated the first question while challenging the second.
Turing turns computation into a general theory
1936: the universal-machine idea
In “On Computable Numbers, with an Application to the Entscheidungsproblem,” published in 1936, Alan Turing described an abstract machine model that helped define what it means for a process to be computable. A universal machine could simulate other machines when supplied with an appropriate description and input. This was not an AI paper in the later disciplinary sense; it was a foundation for theoretical computer science and general-purpose computation.
Rank #3
Babbage had designed a machine that could follow different programs. Turing supplied a mathematical account of a machine capable of representing and executing any computation within the model. His work connected flexible instructions to a precise theory.
1940s: from abstraction to electronic hardware
Turing also worked on the Automatic Computing Engine, or ACE, an electronic stored-program computer design. The Pilot ACE was completed in 1950 and became one of the early digital programmable computers. This work helped bridge the gap between mathematical descriptions of computation and practical electronic machines. The Science Museum’s history links the ACE project with Babbage, Lovelace and Turing.
When Turing made machine intelligence a testable question
In “Computing Machinery and Intelligence” (1950), Turing asked whether machines could think, then replaced that difficult question with the imitation game. In its familiar form, a human judge communicates through text and tries to distinguish a machine from a human participant.
Rank #4
The test is behavioral. It evaluates whether a system can produce conversation that is difficult to distinguish from a person’s, not whether the system is conscious, truthful, reliable, deep-thinking or generally competent. Turing used the reformulation to make debate more operational, not to provide a universal measurement of intelligence. A system can sound persuasive while making factual, logical or procedural errors.
Turing answers Lovelace
Turing directly discussed what became known as Lady Lovelace’s objection. He suggested that a machine could be trained, much as a child is educated, and could produce behavior not anticipated in every detail by its programmer. Learning therefore offered a way to challenge the idea that following an initial program necessarily limits a machine to outcomes explicitly foreseen by its designer. This was an argument about possibility, not evidence that contemporary machines possess human understanding.
When did artificial intelligence begin?
The answer depends on what “begin” means.
- Intellectual antecedents: Babbage’s programmable-machine designs and Lovelace’s 1843 account of symbolic computation.
- Theoretical foundations: Turing’s 1936 work on computability and related developments in logic and mathematics.
- A named research field: the 1955 Dartmouth proposal and its 1956 workshop, conventionally treated as AI’s founding event.
Dartmouth did not create every idea later associated with AI. It gave an emerging collection of problems a shared name, institutional setting and research agenda. The term “artificial intelligence” belongs to that mid-twentieth-century program, not to Babbage or Lovelace’s nineteenth-century projects. Historical discussion of the naming and field formation appears in this Oxford Academic study.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Best Value
A concise chronology
| Date | Development | Significance |
|---|---|---|
| June 14, 1822 | Babbage presented the Difference Engine concept to the Royal Astronomical Society. | Public beginning of his calculating-machine project. |
| 1830s | Babbage developed the Analytical Engine design. | Move toward programmable general-purpose computation. |
| 1833 | Ada Byron met Babbage. | Beginning of their important intellectual relationship. |
| 1843 | Lovelace published her translation and notes on Menabrea’s description. | Major early account of programmable symbolic computation. |
| 1843 | Note G described a Bernoulli-number procedure. | Widely regarded as the first published computer program. |
| 1936 | Turing published “On Computable Numbers.” | Formal foundation for universal computation. |
| 1945–1946 | Turing developed the ACE design and report. | Bridge between mathematical computation and electronic architecture. |
| 1950 | Turing published “Computing Machinery and Intelligence.” | Early direct treatment of machine intelligence and learning machines. |
| 1955 | The Dartmouth proposal used the term “artificial intelligence.” | Naming and framing of a new research program. |
| 1956 | The Dartmouth workshop took place. | Conventional founding milestone of AI as a field. |
What this history gets right—and what it does not
Ideas they anticipated
- Programmable machines rather than single-purpose calculators.
- Instructions and representations that can operate on symbols.
- General-purpose computation.
- Learning as a possible source of behavior not specified in advance.
- Philosophical disputes over originality, rules and intelligence.
Technologies they could not have foreseen
- Transistors, integrated circuits and modern digital memory.
- Internet-scale data and cloud computing.
- Neural networks trained on massive datasets.
- Generative language models, GPUs and contemporary benchmarks.
There was no straight pipeline from Babbage’s workshop to today’s AI. Mechanical engineering limits, funding, institutional change and the later contributions of formal logic, electronics, statistics, cybernetics, information theory and neuroscience all mattered. Lovelace’s notes also did not remain continuously influential through the nineteenth century; their recognition grew substantially later.
The historical answer to “Who invented AI?”
No single person invented artificial intelligence. Babbage supplied an early design for programmable machinery. Lovelace explained its general symbolic potential while insisting that the engine did not originate ideas on its own. Turing formalized universal computation, connected it to electronic computer design and made machine intelligence a question that could be discussed through observable behavior and learning.
AI emerged later from many traditions. The most useful historical lineage is therefore a chain of changing questions: Can calculation be automated? Can one machine follow many programs? Can symbols stand for more than numbers? Can a machine learn? And is intelligent behavior something we judge from what a system does rather than from an inaccessible inner essence?
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
