This document outlines a framework for evaluating Artificial General Intelligence (AGI) by adopting and adapting the Cattell-Horn-Carroll (CHC) theory of human intelligence. The framework decomposes general intelligence into ten core cognitive components (broad abilities) and numerous narrow cognitive abilities. Solving all the tasks corresponding to these abilities implies an AGI Score of 100%. Here is a comprehensive list of each cognitive ability.

General Knowledge (K): Knowledge that is familiar to most well-educated people or is important enough that most of them have been exposed to it. • Commonsense: The vast set of shared, obvious background knowledge about how the world works. • Science: Knowledge of the natural and physical sciences. • Social Science: Understanding of human behavior, societies, and institutions. • History: Knowledge of past events and objects. • Culture: Cultural literacy and awareness.

Reading and Writing Ability (RW): Captures all of the declarative knowledge and procedural skills a person uses to consume and produce written language. • Letter-Word Ability: The ability to recognize letters and decode words. • Reading Comprehension: The ability to understand connected discourse during reading. • Writing Ability: The ability to write with clarity of thought, organization, and good sentence structure. • English Usage Knowledge: Knowledge of writing in the English language with respect to capitalization, punctuation, usage, and spelling.

Mathematical Ability (M): The depth and breadth of mathematical knowledge and skills. • Arithmetic: The manipulation of numbers using basic operations and solving word problems. • Algebra: The study of symbols and the rules for combining them to express general relationships and solve equations. • Geometry: The study of shapes, space, size, position, and distance. • Probability: The quantification of uncertainty by assigning numbers from 0 to 1 to events. • Calculus: The mathematics of change and accumulation.

On-the-Spot Reasoning (R): The deliberate but flexible control of attention to solve novel "on the spot" problems that cannot be performed by relying exclusively on previously learned habits, schemas, and scripts. • Deduction: Reasoning from general statements or premises to reach a logically guaranteed conclusion. • Induction: Discovering the underlying principles or rules that determine a phenomenon's behavior. • Theory of Mind: Attributing mental states to others and understanding how those states may differ from one's own. • Planning: Devising a sequence of actions to achieve a specific goal. • Adaptation: The ability to infer unstated classification rules from a simple performance feedback sequence.

Working Memory (WM): The ability to maintain, manipulate, and update information in active attention. (Often referred to as short-term memory.) • Textual Working Memory: The ability to hold and manipulate sequences of verbal information presented textually. – Recall: Remember a short sequence of elements (digits, letters, words, nonsense words) and answer basic questions about them. – Transformation Sequence: Remember and update a short list of digits following a sequence of operations. • Auditory Working Memory: The ability to hold and manipulate auditory information, including speech, sounds, and music. – Recall: Remember a collection of voices, utterances, and sound effects and answer basic questions about them. – Transformation Sequence: Remember and modify a short utterance with a variety of transformations. • Visual Working Memory: The ability to hold and manipulate visual information, including images, scenes, spatial layouts, and video. – Recall: Remember a collection of images and answer basic questions about them. – Transformation Sequence: Transform a visual input following a sequence of operations. – Spatial Navigation Memory: Represent a sense of location in an environment. – Long Video Q&A: Watch a long video or a movie and answer basic questions about it. • Cross-Modal Working Memory: The ability to maintain and modify information presented across different modalities.

Long-Term Memory Storage (MS): The ability to stably acquire, consolidate, and store new information from recent experiences. • Associative Memory: The ability to link previously unrelated pieces of information. – Cross-Modal Association: Form a link between two previously unrelated stimuli, such that presenting one later activates recall of the other. – Personalization Adherence: Associate specific rules, preferences, or corrections with a distinct interaction context and apply them consistently and unprompted over time. – Procedural Association: Acquire and retain a sequence of associated steps or rules (a procedure) and execute them when cued with the procedure's name. • Meaningful Memory: The ability to remember narratives and other forms of semantically related information. – Story Recall: Remember the gist of stories. – Movie Recall: Remember the gist of movies. – Episodic Context Recall: Remember specific events or experiences, including their context (the "what, where, when, and how"). • Verbatim Memory: The ability to recall information exactly as it was presented, requiring precise encoding of specific sequences, sets, or designs, often independent of meaning. – Short Sequence Recall: Exactly recall short sequences of text after a delay. – Set Recall: Recall a set (order of recall does not matter). – Design Recall: Recall the spatial arrangement and structure of visual information.

Long-Term Memory Retrieval (MR): The fluency and precision with which individuals can access long-term memory. • Retrieval Fluency (Fluency): The speed and ease of generating ideas, associations, and solutions based on stored knowledge. – Ideational Fluency: Rapidly produce a series of ideas, words, or phrases related to a specific condition, category, or object. – Expressional Fluency: Rapidly think of different ways of expressing an idea. – Alternative Solution Fluency: Rapidly think of several alternative solutions to a practical problem. – Word Fluency: Rapidly produce words that share a non-semantic feature. – Naming Facility: Rapidly call common objects by their names. – Figural Fluency: Rapidly draw or sketch as many things as possible. • Retrieval Precision (Hallucinations): The accuracy of accessed knowledge, including the critical ability to avoid confabulation (hallucinations).

Visual Processing (V): The ability to analyze and generate natural and unnatural images and videos. • Perception: The ability to accurately interpret and understand visual input. – Image recognition: Classify images of commonplace objects, places, or facial expressions, including distorted images. – Image captioning: Generate a concise, human-like text description for the visual content of an image. – Image anomaly detection: Detect whether something is anomalous in an image, or what is missing from an object. – Clip captioning: Generate a concise, human-like text description of a short video segment. – Video anomaly detection: Detect whether a short video segment is anomalous or implausible. • Visual Generation: The ability to synthesize images and short videos. • Visual Reasoning: The ability to solve problems and make inferences using spatial logic and visual abstractions. • Spatial Scanning: The speed and accuracy of visually exploring a complex field.

Auditory Processing (A): The ability to discriminate, remember, reason, and work creatively on auditory stimuli, which may consist of tones and speech units. • Phonetic Coding: The ability to hear phonemes distinctly, blend sounds into words, and segment words into parts, sounds, or phonemes. • Speech Recognition: The ability to transcribe a spoken audio signal into its corresponding sequence of text. • Voice: The quality and responsiveness of the AI's synthesized voice. – Natural speech: Utter sentences or paragraphs that sound natural and not robotic. – Natural conversation: Maintain conversational fluidity without long delays or excessive interruptions. • Rhythmic Ability: The ability to recognize and maintain a musical beat, including reproducing rhythms, detecting differences between rhythms, and synchronizing by playing or humming along. • Musical Judgment: The ability to discriminate and judge simple patterns in music.

Speed (S): The ability to perform simple cognitive tasks quickly. • Perceptual Speed–Search: The speed of scanning a visual or textual field to find specific targets. • Perceptual Speed–Compare: The speed of comparing two or more stimuli to identify similarities or differences. • Reading Speed: The rate at which text can be processed with full comprehension. • Writing Speed: The rate at which text can be generated or copied. • Number Facility: The speed and accuracy of performing basic arithmetic operations. • Simple Reaction Time: The time taken to respond to a single, anticipated stimulus. • Choice Reaction Time: The time taken to respond correctly when presented with one of several possible stimuli. • Inspection Time: The speed at which subtle differences between visual or auditory stimuli can be perceived. • Comparison Speed: The time taken to make a judgment comparing two stimuli on a specific attribute. • Pointer Fluency: The speed and accuracy of moving a pointer, such as a virtual mouse.

Figure 2 summarizes the broad and narrow capabilities tested. The ten core cognitive components of our AGI definition: General Knowledge (K), Reading & Writing (RW), Mathematical (M), On-the-Spot Reasoning (R), Working Memory (WM), Long-Term Memory Storage (MS), Long-Term Memory Retrieval (MR), Visual (V), Auditory (A), Speed (S), each further decomposed into narrow abilities such as Commonsense, Letter-Word, Arithmetic, Deduction, Textual Working Memory, Associative Memory, Fluency, Perception, Phonetic Coding, Perceptual Speed, etc.