Artificial General Intelligence
Also known as AGI
Overview
The field asks what it means for an AI system to be generally capable, rather than merely strong on a narrow task, and how such capability could be built and measured. Approaches include scaling foundation models, coordinating specialized agents, adding explicit memory and planning, and drawing on cognitive science or formal theories of learning. Evaluations use multitask benchmarks, games, and interactive environments to test transfer, reasoning, and long-term performance, but scores can be distorted by benchmark design, tools, and specialization. A central tension is whether generality emerges from a common underlying ability or requires many specialized skills, alongside unresolved questions about feasibility, safety, and who shapes AGI’s development.
Latest papers
24 papers3 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.