Google Deepmind hackathon to pit meatbags v machines

"DeepMind wants to tighten up that definition. The team at the Google-owned AI research and development shop reported this week that they had developed a "cognitive taxonomy" for measuring the tech industry's progress toward universally useful AGI, along with a three-stage test for benchmarking AI system performance against human capabilities."

"Run AI models and humans through the same cognitive benchmarks, say DeepMinders, and you'll get a good estimate of when a single AI is able to meet or exceed human capabilities on all ten areas of its taxonomy, which are divided into two main areas, as the team describes in a paper."

"First up are eight basic building blocks of human cognition, which have been previously defined by other researchers: Perception, generation, attention, learning, memory, metacognition, and executive functions."

Google DeepMind has created a scientific framework to measure progress toward artificial general intelligence (AGI) by establishing a cognitive taxonomy and three-stage performance test. The framework compares AI models and humans across the same cognitive benchmarks to determine when AI meets or exceeds human capabilities. The taxonomy includes eight fundamental building blocks of human cognition—perception, generation, attention, learning, memory, metacognition, and executive functions—plus two additional areas. This approach aims to provide empirical, grounded measurement of AGI progress, addressing the problem of loosely defined terminology in the AI industry where the definition of AI has shifted as machine learning applications became more prevalent.

Read at Theregister