CASRAI Dictionary

Tag: artificial intelligence

  • What Is Artificial Intelligence? Definition and History

    Artificial intelligence (AI) is the branch of computer science concerned with building systems that perform tasks normally requiring human intelligence, such as perception, reasoning, language understanding and decision-making. As a research field it spans both symbolic approaches, which encode knowledge and rules explicitly, and statistical approaches, which learn patterns from data. For the research community, AI is best understood not as a single technology but as a long-standing discipline with a measurable history, contested definitions and evolving documentation standards.

    A working definition of artificial intelligence

    There is no universally agreed definition of artificial intelligence, partly because the goalposts move: tasks once considered to require intelligence, such as optical character recognition, become routine engineering and stop being called AI. A durable, standards-friendly definition treats AI as the study and construction of agents that perceive their environment and take actions to maximise a defined objective. This framing accommodates everything from a rule-based expert system to a modern neural network without privileging any one method.

    Because the term is so elastic, research-standards bodies encourage authors to describe the specific method used, rather than the marketing label. A paper that says it “used AI” tells a reader very little; one that names the model class, training data and evaluation protocol is reproducible. The casrai.org research dictionary exists precisely to stabilise this vocabulary across disciplines.

    Narrow AI versus general AI

    Almost all systems deployed today are examples of narrow AI (also called weak AI): they are built for a single, bounded task such as translating text, recommending content or classifying images. A narrow system that excels at one task has no capacity to transfer that competence to another domain.

    Artificial general intelligence (AGI) refers to a hypothetical system with the broad, flexible competence of a human across arbitrary tasks. AGI remains a research aspiration rather than an existing artefact, and claims of its arrival should be treated with scholarly caution. Keeping the narrow/general distinction explicit prevents the overstatement that often clouds reporting on AI in research outputs.

    Symbolic AI versus statistical approaches

    The field has long been organised around two broad paradigms. Symbolic AI (sometimes called “good old-fashioned AI”) represents knowledge as symbols and manipulates them with explicit logical rules; expert systems and classical search and planning belong here. Its strengths are transparency and the ability to explain a decision step by step.

    Statistical or machine-learning approaches instead infer behaviour from data. Rather than hand-coding rules, an engineer specifies a model and an objective, and the system learns parameters that fit observed examples. This paradigm now dominates practical AI, and it underpins the techniques discussed in our companion piece on machine learning concepts and methods. The two paradigms are increasingly combined in neuro-symbolic systems that pair learned perception with explicit reasoning.

    A brief history: Dartmouth 1956 to the deep-learning era

    The field was named at the Dartmouth Summer Research Project on Artificial Intelligence in 1956, a workshop organised by John McCarthy, Marvin Minsky, Nathaniel Rochester and Claude Shannon. Early optimism produced symbolic reasoning programs and the first neural-network models, but progress stalled when problems proved harder than expected, producing the so-called AI winters of reduced funding and interest in the 1970s and again in the late 1980s.

    The modern resurgence, often dated to the early 2010s, came from the convergence of large datasets, graphics-processing hardware and improved training methods, ushering in the deep-learning era. These advances are explored further in our overview of neural networks and deep learning, and they set the stage for today’s generative models.

    Period Milestone Significance
    1950 Turing’s “Computing Machinery and Intelligence” Proposed the imitation game (Turing test)
    1956 Dartmouth workshop Coined the term “artificial intelligence”
    1970s, late 1980s AI winters Funding and interest contracted
    2010s Deep-learning breakthroughs Data plus GPUs revived neural networks

    The Turing test

    In 1950 Alan Turing proposed what is now called the Turing test: rather than asking whether a machine can “think”, he asked whether a human interrogator, conversing by text, could reliably distinguish the machine from a person. The test reframed an unanswerable philosophical question as an operational one. It remains a touchstone for discussion, though contemporary researchers treat it as a thought experiment rather than a benchmark of genuine understanding, and it does not measure reasoning, safety or factual accuracy.

    Why definitions matter for the research record

    Precise terminology is not pedantry; it is the foundation of reproducibility and credit. When AI methods feature in a study, readers and reviewers need to know exactly what was done. This connects to broader work on contribution transparency captured by the CRediT taxonomy and to the emerging disclosure norms tracked in our AI and ML research outputs coverage.

    Frequently asked questions

    Is artificial intelligence the same as machine learning?

    No. Machine learning is a subfield of artificial intelligence concerned with learning from data. AI is the broader discipline and also includes symbolic reasoning, search and planning that need not learn at all.

    Does any current system count as general AI?

    No. All systems in production are narrow AI, built for specific tasks. Artificial general intelligence remains a research aspiration, and claims of its existence should be treated sceptically.

    What was the significance of the 1956 Dartmouth workshop?

    It is where the term “artificial intelligence” was coined and where the field was effectively founded as a distinct research discipline, setting a shared agenda for the decades that followed.

    Does passing the Turing test prove a machine is intelligent?

    Not in any deep sense. The test measures whether a machine can imitate human conversation convincingly, not whether it understands, reasons soundly or is factually reliable.

  • Natural Language Processing (NLP) in Research: A Plain Guide

    Natural language processing (NLP) is the field of artificial intelligence concerned with making human language machine-processable, so computers can read, interpret, generate and respond to text and speech. It combines linguistics, statistics and machine learning to turn unstructured language into structured signals a model can work with. NLP now underpins search engines, translation tools, literature-screening systems and the large language models behind modern research assistants.

    From raw text to numbers

    Computers operate on numbers, not words, so the first job of any NLP pipeline is to convert language into a numerical form. Two steps dominate this process.

    Tokenisation splits text into smaller units called tokens, which may be words, sub-words or characters. Modern systems favour sub-word tokenisation because it handles rare words and morphology gracefully without an unmanageably large vocabulary.

    Embeddings then map each token to a dense vector of numbers, positioning words with similar meanings near one another in a high-dimensional space. Because embeddings capture semantic relationships learned from large text corpora, “clinician” and “physician” sit close together while “clinician” and “granite” do not. This numerical representation is what downstream models actually learn from. The reliance on learned representations connects NLP to the wider field, which we introduce in what is machine learning.

    Transformers: the architecture that changed NLP

    The transformer, introduced in 2017, is the architecture behind most current NLP systems. Its key innovation is the attention mechanism, which lets the model weigh the relevance of every word to every other word in a sequence, regardless of distance. This captures long-range context that earlier sequential models struggled with, and it parallelises well, enabling training on vast corpora. Large language models are transformers scaled to billions of parameters and trained on enormous text collections.

    Common NLP tasks

    NLP is best understood through the tasks it performs. The table below lists those most relevant to research.

    Task What it does Research example
    Text classification Assigns a category to a document Screening abstracts for a systematic review
    Named entity recognition Identifies entities such as genes, drugs or places Extracting chemical names from papers
    Machine translation Converts text between languages Reading non-English literature
    Summarisation Condenses long text into key points Digesting large document collections
    Question answering Returns answers from a body of text Querying a corpus of protocols

    How researchers use NLP

    Across disciplines, NLP accelerates work that would be impractical by hand. Systematic reviewers use classification to triage thousands of abstracts. Biomedical teams use named entity recognition to mine entities from the literature at scale. Social scientists apply topic modelling and sentiment analysis to large text archives. Curators and metadata specialists increasingly use NLP to normalise terminology against controlled vocabularies such as the CASRAI dictionary, improving the consistency of research records.

    Caveats and reproducibility concerns

    NLP systems inherit the limitations of their training data. Models can encode and amplify bias present in source corpora; they can produce fluent but factually wrong output, often called hallucination; and their behaviour can shift when an underlying model is updated. For research use, these issues raise real reproducibility questions: a result obtained from one model version may not replicate on the next. Documenting the exact model, version, prompt and preprocessing is therefore essential, a theme we explore in our coverage of reproducibility of machine learning research and our broader AI and ML research outputs hub. Treating NLP as a tool whose outputs require human verification, not an oracle, keeps it trustworthy.

    Frequently asked questions

    What is the difference between NLP and machine learning?

    Machine learning is the general study of systems that learn patterns from data. NLP is the application of those techniques, among others, specifically to human language. Most modern NLP is built on machine learning, but they are not the same thing.

    What are embeddings in simple terms?

    Embeddings are lists of numbers that represent the meaning of a word or piece of text, arranged so that similar meanings have similar numbers. They let a model treat “begin” and “start” as related while keeping unrelated words apart.

    Why are transformers so important in NLP?

    Transformers use an attention mechanism to weigh the relevance of all words in a sequence at once, capturing long-range context and training efficiently at scale. They are the foundation of nearly all current large language models.

    Can I trust NLP output in research?

    Only with verification. NLP models can be biased, can fabricate plausible-sounding content, and can change between versions. Record the model, version and settings, and check outputs against authoritative sources, as set out in our guidance for authors.