From Alan Turing's pioneering vision to modern large language models — explore the key milestones that shaped artificial intelligence.
Alan Turing publishes «On Computable Numbers», introducing the concept of a universal machine — the theoretical foundation of all modern computers. This paper lays the groundwork for thinking about machines that can simulate any human reasoning process.
FoundationTuring poses the question «Can machines think?» and proposes the Imitation Game — later known as the Turing Test. A machine passes if a human interrogator cannot reliably distinguish it from a human. This remains a benchmark for AI to this day.
PhilosophyJohn McCarthy, Marvin Minsky, Nathaniel Rochester, and Claude Shannon coin the term «Artificial Intelligence» at the Dartmouth Summer Research Project. This event is widely considered the official birth of AI as an academic discipline.
Birth of AIJoseph Weizenbaum at MIT creates ELIZA, a natural language processing program that simulates a Rogerian psychotherapist. Despite being a simple pattern-matching system, users formed emotional attachments, revealing the power of human-AI interaction.
NLPDavid Rumelhart, Geoffrey Hinton, and Ronald Williams publish the backpropagation algorithm, making it practical to train multi-layer neural networks. This breakthrough reignites interest in connectionist approaches and deep learning.
Deep LearningIBM's Deep Blue becomes the first computer to defeat a reigning world chess champion, Garry Kasparov, in a match under standard tournament conditions. The victory captures global imagination and demonstrates the potential of brute-force search combined with specialized hardware.
MilestoneFei-Fei Li releases ImageNet — a massive dataset of 14 million labeled images. In 2012, Alex Krizhevsky's CNN (AlexNet) achieves a dramatic accuracy leap in the ImageNet competition, sparking the deep learning revolution.
Computer VisionIan Goodfellow introduces Generative Adversarial Networks (GANs), where two neural networks compete: a generator creates synthetic data, and a discriminator tries to detect fakes. This framework becomes the foundation of modern generative AI.
Generative AIGoogle researchers publish «Attention Is All You Need», introducing the Transformer architecture. The self-attention mechanism revolutionizes natural language processing, enabling parallel processing and long-range dependencies far better than RNNs or LSTMs.
ArchitectureOpenAI releases GPT-3 with 175 billion parameters, demonstrating emergent abilities like few-shot learning, translation, code generation, and creative writing. It shows that scaling language models leads to qualitatively new capabilities not present in smaller models.
LLMOpenAI launches ChatGPT based on GPT-3.5 and GPT-4, making conversational AI accessible to hundreds of millions of users worldwide. It becomes the fastest-growing consumer application in history, reaching 100 million users in just two months.
MainstreamAI systems gain multimodal capabilities (text, image, audio, video), real-time reasoning, and tool use. Models like Claude, Gemini, and GPT-4o demonstrate human-level performance on complex benchmarks. AI agents autonomously browse the web, write code, analyze data, and orchestrate workflows — ushering in the age of agentic AI.
Present