Generative AI
Generative AI is software that produces new content, text, images, audio, code, by sampling patterns learned from vast training data, rather than only classifying or retrieving what already exists.
No jargon, no math
In plain words, what is it?
Classic AI sorts incoming mail into folders; generative AI writes the reply. A generative model trains on billions of examples and learns the probability distribution of what comes next: given this context, which word, pixel or note is most likely. At generation time you give it a prompt, it converts the prompt into a numeric state, then samples one small piece at a time (a token of text, a block of pixels), feeds that piece back in, and repeats until the output is complete. Nothing is copied from a database: the output is recomposed from the learned pattern, different every time. That is also where it breaks. The model predicts what is plausible, not what is true, so left unconnected to your data it will invent figures with total confidence.
Terms that live nearby.
Large Language Model
A large language model (LLM) is an AI system trained on massive collections of text to predict the next token, which gives it a general ability to write, summarise, translate and reason without being programmed for each task.
ReadTransformer
The transformer is the neural network architecture, published by Google researchers in 2017, behind every modern language model: it processes all tokens of a text at once and uses attention to let each token weigh the relevance of every other token.
ReadDiffusion Model
A diffusion model is a generative model trained to reverse a gradual noising process: it learns to remove noise step by step, so at generation time it can start from pure static and refine it into an image, sound or other output that matches your prompt.
Read§ 03
Want this running in your stack?
We design and ship AI systems with permissions, logs and human approval. It starts with a 30-minute scoping call.