New cognitive framework proposed to measure progress toward AGI

A new paper, "Measuring Progress Toward AGI: A Cognitive Taxonomy," sets out a scientific foundation for assessing AI systems' cognitive capabilities, drawing on decades of research from psychology, neuroscience and cognitive science. It defines ten cognitive abilities hypothesized to be important for general intelligence: perception, generation, attention, learning, memory, reasoning, metacognition, executive functions, problem solving and social cognition.

To benchmark systems, the authors propose a three-stage protocol: evaluate AI across a broad suite of cognitive tasks using held-out test sets to avoid data contamination; collect human baselines from a demographically representative sample of adults; and map each AI system’s performance relative to the distribution of human performance for each ability.

agi, cognitive taxonomy, general intelligence, reasoning, metacognition, learning, perception, attention, human baselines, held-out tests