Last reviewed:
What is artificial intelligence (AI)? Simple definition and examples
Artificial intelligence (AI) refers to the set of techniques that enable a machine to perform tasks associated with human intelligence: recognising an image, understanding a text, predicting, deciding, writing. Under EU law, an “AI system” is identified by its capacity to infer: it infers from the input it receives how to generate predictions, content, recommendations or decisions (AI Act, Article 3).
The term appears in the proposal for the “Dartmouth Summer Research Project on Artificial Intelligence”, dated August 31, 1955 and signed by John McCarthy, Marvin Minsky, Nathaniel Rochester and Claude Shannon. It planned a two-month, ten-person study in summer 1956, based on the conjecture that every aspect of intelligence can be described precisely enough for a machine to simulate it. France set an official definition in its Journal officiel of December 9, 2018: an “interdisciplinary field” that aims to understand the mechanisms of cognition and imitate them with hardware and software. Two broad families coexist. Symbolic AI applies rules and knowledge written by experts (expert systems, rule engines). Learning-based AI, or machine learning, derives its rules from data; deep learning is the branch based on deep neural networks, and generative AI (ChatGPT, Claude, Mistral) stems from it. For a business, the legal definition matters more. Article 3(1) of the EU AI Act covers a machine-based system with some level of autonomy that “infers, from the input it receives”, how to generate outputs that can influence its environment. Recital 12 calls for alignment with international work: the definition follows the one the OECD revised in November 2023. It excludes systems based on rules defined solely by natural persons to automatically execute operations. On February 6, 2025, the European Commission published non-binding guidelines to help determine whether a system qualifies.
Concrete example
Illustrative case: a French trading SME with 120 employees needs to list its tools to prepare for the AI Act. It finds four. Its email spam filter and the sales forecasting module of its ERP learn from data: they are AI systems within the meaning of Article 3. The writing assistant used by sales staff runs on a generative AI model: it is also an AI system, subject in some cases to the transparency obligations of Article 50. The spreadsheet that calculates discounts using thresholds set by the sales director applies rules written by people: it falls outside the definition. The exercise takes half a day and gives management a clear map of what the regulation covers.
Comparison
| Symbolic AI (rules) | Classic machine learning | Deep learning | Generative AI | |
|---|---|---|---|---|
| Principle | Rules and knowledge written by experts | The model learns rules from data | Neural networks with many layers | Large models that produce new content |
| Business example | Pricing rule engine, expert system | Customer scoring, sales forecasting | Camera-based quality control, speech recognition | Writing assistant, document summaries |
| Data required | Little, but formalised expertise | Structured data, from a few thousand rows up | Large volumes, often unstructured (images, audio, text) | Model pre-trained by a provider, plus your documents |
| Explainability | High: every rule is readable | Medium: depends on the model | Low: hard to explain how it works | Low: variable answers, risk of hallucination |
| Boom period | 1970s-1980s (expert systems) | 1990s-2000s | Since 2012 | Since 2022 (ChatGPT) |
FAQ
What is artificial intelligence, in simple terms?
It is a set of techniques that allow a computer to perform tasks associated with human intelligence: recognising an image, understanding a question, forecasting sales, writing a text. Most current AI learns these tasks from examples rather than following hand-written rules.
What is the definition of artificial intelligence?
France's official definition (Journal officiel, December 9, 2018) describes an interdisciplinary field that studies the mechanisms of cognition and imitates them with hardware and software. The EU legal definition (AI Act, Article 3) covers a machine-based system that infers from its input how to generate predictions, content, recommendations or decisions.
What are everyday examples of artificial intelligence?
Your email spam filter, facial recognition on your phone, machine translation, recommendations on a shopping site, route planning that accounts for traffic, and conversational assistants such as ChatGPT, Claude or Mistral's Le Chat.
Who invented artificial intelligence?
No single person. The term appears in the Dartmouth workshop proposal (August 31, 1955), signed by John McCarthy, Marvin Minsky, Nathaniel Rochester and Claude Shannon; the workshop took place in summer 1956. Work on artificial neurons dates back to 1943 (McCulloch and Pitts).
What is the difference between AI, machine learning and deep learning?
They are nested circles. AI is the overall field. Machine learning is the family of AI that learns from data. Deep learning is the part of machine learning that uses neural networks with many layers. Generative AI, such as ChatGPT, is built on deep learning.
What is generative artificial intelligence?
It is AI that produces new content (text, images, sound, code) from an instruction, instead of only classifying or predicting. It relies on large deep learning models trained on huge volumes of data. ChatGPT, Claude and Midjourney are examples.
See also
Further reading
Explanatory memorandum on the updated OECD definition of an AI system, OECD, March 5, 2024
Sources
- Regulation (EU) 2024/1689 on artificial intelligence (AI Act), Article 3, definitions (official text, AI Act Service Desk, European Commission). https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-3
- A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence, McCarthy, Minsky, Rochester and Shannon, August 31, 1955. http://jmc.stanford.edu/articles/dartmouth/dartmouth.pdf
- Vocabulaire de l'intelligence artificielle (official French AI terminology), Journal officiel No. 0285, December 9, 2018. https://www.legifrance.gouv.fr/jorf/id/JORFTEXT000037783813
- Commission publishes the Guidelines on AI system definition to facilitate the first AI Act's rules application, European Commission, February 6, 2025. https://digital-strategy.ec.europa.eu/en/library/commission-publishes-guidelines-ai-system-definition-facilitate-first-ai-acts-rules-application