A large AI model trained on broad data that can be adapted to many tasks, such as writing, coding, or image analysis.
The term foundation model was popularized by Stanford researchers in 2021 to describe models trained on very large and varied datasets, which then serve as a base for many applications. Large language models are the best-known type, but foundation models also exist for images, audio, video, and scientific data.
Rather than training a new model for each task, a company can start from a foundation model and adapt it through prompting, retrieval, or fine-tuning. For example, a legal software vendor might use a general foundation model and add its own clause library so the model drafts contract summaries in the firm's style. Because many products depend on a small number of foundation models, their strengths, biases, and limitations tend to spread to everything built on top of them.
Further training a pre-trained AI model on a smaller, specific dataset so it performs better on a particular task or style.
Generative AIAI systems that create new content, such as text, images, audio, video, or code, based on patterns learned from training data.
Large Language ModelAn AI model trained on vast amounts of text to understand and generate human-like language, code, and other content.
Open-Weight ModelAn AI model whose trained parameters, or weights, are published so anyone can download, run, and adapt it, subject to its licence.
AI chatbots can help you prepare for a doctor visit, but research shows they are unreliable for self-diagnosis. Here is how to use them safely and the red flags that mean get care now.
AI is moving from pilot projects into everyday clinical work. Here is where it genuinely helps in diagnostics, documentation and triage, and how clinics can adopt it safely.
Where AI actually pays off in logistics and supply chain, from demand planning to routing and warehouses, and the data you need before each one works.