Machine learning articles from across Nature Portfolio

The development of clinically relevant artificial intelligence (AI) models has traditionally required access to extensive labelled datasets, which inevitably centre AI advances around large centres and private corporations. Data availability has also dictated the development of AI applications: most studies focus on common cancer types, and leave rare diseases behind. However, this paradigm is changing with the advent of foundation models, which enable the training of more powerful and robust AI systems using much smaller datasets.

Large language models (LLMs) are becoming less overtly racist, but respond negatively to text in African American English. Such ‘covert’ racism could harm speakers of this dialect when LLMs are used for decision-making.

Artificial neural networks become incapable of mastering new skills when they learn them one after the other. Researchers have only scratched the surface of why this phenomenon occurs — and how it can be fixed.

Increasing the binding affinity of an antibody to its target antigen is key for antibody therapeutics. Here the authors report a pretrainable geometric graph neural network, GearBind, and explore its potential in in silico antibody affinity maturation.

The authors present DPAD, a deep learning method, for dynamical neural–behavioral modeling. It dissociates behaviorally relevant neural dynamics, better predicts neural–behavioral data and reveals insight into where their nonlinearities can be isolated.

The development of clinically relevant artificial intelligence (AI) models has traditionally required access to extensive labelled datasets, which inevitably centre AI advances around large centres and private corporations. Data availability has also dictated the development of AI applications: most studies focus on common cancer types, and leave rare diseases behind. However, this paradigm is changing with the advent of foundation models, which enable the training of more powerful and robust AI systems using much smaller datasets.

As artificial intelligence becomes increasingly integral to the world outside academia, universities face a crucial choice: to use or not to use.

Generative artificial-intelligence tools have been widely adopted across academia, but users might not be aware of all their inherent risks.

Nuance is often lost on artificial-intelligence models, but here’s why society must navigate these grey areas.

The large language model does everything from reading the literature to writing and reviewing its own papers, but it has limited applicability.

Researchers developed an AI-enabled, battery-operated tool that can be operated by clinicians with no sonography experience — and that measures gestational age as accurately as high-specification ultrasound.